Energy system changes in the agricultural sector: testing a value-congruent communication approach to foster positive attitudes

Multiple sectors, including agriculture, are subject to decarbonization and wide-ranging system changes. With a growing distance between agriculture and the public, communicating changes and fostering public acceptance as crucial factors for the success of these changes has become challenging. Therefore, understanding public attitudes towards emerging system changes in unfamiliar sectors helps identify appropriate communication strategies and provides information about possible responses from different stakeholders. To address the scarcity of studies on system changes in agriculture and their acceptability, a multidimensional evaluation of attitudes was undertaken using a factorial survey to test the utility of value-congruent information for enhancing positive attitudes. Electrical field cultivation (EFC) served as a case example for one potential energy system change in agriculture. The data were collected via an open scientific survey panel in Germany (N = 300) using a written vignette in the form of a newspaper article; different benefit foci and hypotheses were tested using path analysis. Although the different benefit foci did not reveal significant differences, problem awareness and knowledge were found to predict positive attitudes toward EFC. For once, results highlight the need to foster overall problem awareness for fossil fuels to garner public support for various pathways. Additionally, results underscore the perceived lack of personal affectedness by transition pathways as a barrier to responding to communication strategies, specifically in unfamiliar sectors such as agriculture. The study outlines the need for stakeholders to design communication strategies under consideration of fostering personal affectedness and overall problem understanding.


Introduction
To meet current concerns about energy security and the shift of power generation from fossil fuels to renewable energy, the expansion of renewable energies and corresponding infrastructure is critical.Recently, the ambitious 2030 target for increasing the share of renewable sources in the European Union's (EU's) overall energy mix was increased from 40% to 45%; achieving this requires all sectors (heating, transport, etc) to uniquely contribute by saving energy and increasing their renewable energy use (European Commission 2022).The agricultural sector is one sector that is vital to a successful energy transition by providing various resources to drive its success while also consuming a significant portion of the world's energy production (Rahman et al 2022).
Sitting at the intersection of energy transition and agriculture, agricultural machine innovations can potentially utilize alternative power supplies, such as electrical power from renewable energy sources (Gorjian et al 2022).However, the successful transition of the sector toward electrification requires the interplay of micro and macro factors, in which public acceptance is crucial (Vogeler et al 2021, Olvermann et al 2023).The social sciences, particularly the field of psychology as an empirical discipline with its core competencies in understanding human behavior and attitudes, provide diverse starting points and high-utility methods for it (Weber 1997, Burch et al 2023).The relevance of public acceptance for energy transition has crystallized from decades of extensive research, especially in environmental psychology; underlying attitudes and positive emotions are linked to acceptance of and support for policies, energy system changes, and associated energy technologies (Steg et al 2015, Abrahamse andSchuitema 2020, Emmerich et al 2020).
To foster acceptance, it is imperative to understand the factors that influence the development of positive attitudes (e.g., acceptability or affect) that lead to acceptance (Steg et al 2015).Few studies have focused on attitudes toward emerging energy technologies, future scenarios, or energy system changes in the early development stage (Abrahamse and Schuitema 2020, Palomo-vélez et al 2021, Baur et al 2022), and little attention has been paid to innovations and their acceptability in the agricultural sector (Specht et al 2016, Pfeiffer et al 2020).However, such investigations are essential to anticipate future public reactions, which can pinpoint risk-laden outcomes (Boudet 2019).Moreover, these future reactions serve to abstract guidelines to design successful transition processes by addressing expected risks and built upon benefits.There is also a growing need for causality research to assess the determinants of acceptability, to support the design of communication strategies and policies for industrial and governmental stakeholders (Busse andSiebert 2018, Ejelöv andNilsson 2020).
The agricultural sector and its energy system changes make a highly interesting case for investigating public responses due to the rising distance between the public and the agricultural sector, which poses a challenge for promoting positive attitudes.The public lacks knowledge about the sector, resulting in an absence of dialogue between the two parties, despite the public having multiple touchpoints with the sector in their everyday live (e.g. as consumers or residents).The consequences are incomprehension of measures and proceedings as well as sweeping criticism by the public (Christoph-Schulz et al 2018, Berkes et al 2020).This marks the need to provide more information about wide-ranging system changes in the sector, including new and emerging agricultural technologies and processes; which is crucial to better engage the public (Steg et al 2015).While understanding public attitudes can provide vital information about possible responses to new technology, system changes, or political measures (Boudet 2019, Ejelöv andNilsson 2020), a common mistake lies in neglecting the fit between the audience and provided information (Nilsson et al 2016, Sargisson et al 2020).
To address this gap, and the lack of studies on future systemic changes in agriculture and public perceptions thereof, this research examined the factors that influence attitudes toward agricultural energy transition pathways and explored how positive attitudes can be promoted.This is attained by drawing on empirical and theoretical insights from environmental psychology around the enhancement and nuanced understanding of attitudes.The object of the study was electrically driven field cultivation (EFC), which serves as a potential agricultural energy system change.First, the relevant antecedents of three dimensions (cognitive, affective, and behavioral) of attitudes toward EFC were identified to obtain a comprehensive picture of the efficacy of a particular communication strategy.Second, a factorial survey design was employed to test whether valuecongruent messaging is a potential communication approach for fostering positive attitudes.By taking this approach, recommendations for the design of the emerging transition pathway can be abstracted early on, along with significant indicators to develop a communication strategy for the public when implementation is due.
The present study contributes to the state of the art by investigating one energy system change (here: EFC) in its early development phase (Abrahamse andSchuitema 2020, Emmerich et al 2020).Additionally, the empirical investigation is located in the agricultural sector as a critical industry for a successful energy transition and therefore combines two noteworthy topic areas for the energy transition: agriculture and electrification.More specifically, we aim to make a case for the psychological discipline and its valuable insights in the agricultural sector, which plays a crucial role in climate change mitigation.On a practical-oriented note, we further believe that this study inherently has the potential to strengthen the dialogue between the public and future developments in the agriculture sector.On a methodological note, one communication strategy by means of a factorial survey design is applied to enhance experimental realism and increase internal and external validity (Busse andSiebert 2018, Ejelöv andNilsson 2020).Additionally, attitudinal responses by the public are captured on multiple dimensions to expand the understanding beyond rare acceptability (see Batel et al 2013).

Study context: EFC at the intersection of energy transition and the agricultural sector
The agricultural sector plays a non-negligible role in energy transition (Sutherland et al 2015).First, it provides land resources in the form of renewable energies (e.g., windmills).Second, farmers are critical actors in energy transition as they increasingly operate as Energiewirte1 (energy farmers) as a secondary or main activity.Third, on one side, the sector is heavily dependent on fossil fuels due to energy-intensive processes (Gorjian et al 2022), but on the other side, it experiences intense problems arising from climate change issues, such as detrimental weather events, which result from emissions (Anderson et al 2020).This underscores the necessity of replacing fossil fuels within the sector (Gorjian et al 2022).
At the agricultural-electrification intersection, there are increasing attempts to emphasize the potentially mutually beneficial relationship, e.g., the sector's role in wind farm development (Bessette and Mills 2021) or applying renewable energies to agricultural activities (Gorjian et al 2022).Generally, electrical vehicle use in agriculture is considered a suitable option for farmers, but few studies have investigated the potential of different alternative energy sources for agricultural machinery (Gorjian et al 2021, Schneider et al 2023) or the perceptions of e-mobility in agriculture by farmers (Riedner et al 2019, Sok andHoestra 2023).
To enrich empirical research on the decarbonization of agriculture, the present study examines one potential transition pathway: electrical field cultivation (EFC; Ghobadpour et al 2019, Gorjian et al 2021).EFC, which is relatively unknown to the public and characterized by the use of new technologies and processes, is highly interesting.The future vision for electrically-powered farming machinery includes various innovations, such as larger storage batteries, medium-sized tractors, or new electrified machine concepts (e.g., drones) to enable the electrification of various field management processes (Gorjian et al 2022).These innovations may have farreaching consequences that go beyond their application to energy systems, towards new power relations between stakeholders in the sector and an extension of the electrical energy infrastructure.Hence, the public will be affected in their roles as residents, taxpayers, political agents, and consumers.Therefore, understanding the drivers of public acceptance of EFC is important for the success of electrically-powered farming.
In this study, Germany was chosen for an in-depth case study due to its high political ambitions in contributing to a successful energy transition and the expansion of renewable energies (Die Bundesregierung 2022).The German government is applying a combination of different political instruments, taking the Renewable Energy Sources Act (EEG) as the driving force (Rogge and Johnstone 2017).Since 2000, the EEG has been continuously modified under stringent monitoring and evaluation (Rogge and Johnstone 2017).The recently passed EEG 2023 outlines extensive independence from fossil fuels and greenhouse gas-free power generation by 2035 (Die Bundesregierung 2022).

Attitudes toward sustainable energy technologies and system changes
Since EFC involves a combination of non-technological (e.g., policy measures) and technological innovations (e.g., machine concepts), we define it as one potential energy system change and therefore draw on the rich literature on attitudes toward energy technologies and policies.
When exploring public acceptance of energy system changes that are not yet implemented, acceptance as a behavior toward technology cannot be assessed.The best predictor remains acceptability as an attitude toward change as well as an attitude toward potential behavior (behavioral intentions) regarding technology (Huijts et al 2012).In line with other researchers, we understand attitudes to be an evaluative judgment of an entity (in this case EFC) that concerns its favorableness (Ajzen 2001, Schuitema et al 2020).Even though a general judgment is of great value to anticipate behavioral actions, statements of being in (dis)favor of energy system changes provide little starting point for designing targeted group-oriented communication and implementation pathways.Therefore, it is necessary to identify how this judgment is informed by various factors, which consequently influence the intention to accept (or not) a certain technology or policy.Drawing on Huijts et al 2012) comprehensive framework, which outlines psychological factors influencing acceptance of sustainable energy technologies, attitudes are indirectly and directly shaped by cognitions (e.g., risk and benefit perception) and affect (emotions and feelings) formed when thinking about technology or associated decisions.Previous empirical studies have repeatedly shown that positive attitudes toward a technology decrease as concerns about the negative consequences of various dimensions of the technology increase.However, perceived benefits have the opposite effect (De Groot et al 2013, Fung et al 2014).Similar effects on attitudes can be observed for positive affect (e.g., pride or joy) and negative affect (e.g., fear and anger) toward energy technologies, which distinctively predict attitudes (Huijts et al 2012, Brosch et al 2014, Cousse et al 2021).Consistent with the traditional understanding of a trinity of attitudes, each cognitive and affective attitude builds on a third component for evaluating sustainable energy technologies-behavioral intention-which is the best proxy for actual future behavior in future-oriented research (Liu et al 2016).One of the most informative behavioral intentions regarding energy technologies, policies, or system changes is the willingness to pay for new technologies, products, or taxes (Danne et al 2020).Building on Huijts et al 2012) framework, we go beyond investigating acceptability as a favorable outcome and add behavioral intention, risk and benefit perceptions, and positive and negative emotions as distinct evaluative responses for EFC.For electrification in an unknown sector such as agriculture, we argue that multicomponent investigation allows for a more informative evaluation by the public.For simplicity, we refer to all six components as attitudinal components (figure 1).

Relevant antecedences for attitudes toward energy technologies and energy system changes
To successfully communicate EFC to the public, it was critical to determine individual predispositions for the attitudinal components.Therefore, Huijts et al (2012) comprehensive framework for sustainable energy technologies provided a range of potential influencing psychological factors.Given the combination of an unknown system (agricultural) and a prominent technological application (electrification), we focused on the public's prior knowledge, problem awareness and value orientations.Moreover, we present value-congruent messaging as a situational factor not only to predict but also to potentially enhance positive attitudes.

Prior knowledge
Within Huijts et al (2012) framework, knowledge about a specific energy system change or technology is listed as both a direct and indirect antecedent for acceptability.Subjective knowledge specifically has been found to influence which information sources are used for sustainable energy technologies (Van Rijnsoever and Farla 2014), which marks it as a notable source for evaluation, whereas empirical findings are mixed when it comes to the favorableness of positive attitudes.Given the novel energy alternative of 'hydrogen energy', a low level of subjective knowledge by the public is linked to feelings of uncertainty that might be pervaded by skepticism and negative attitudes (Gordon et al 2022).Regarding energy system changes, a recent study by Linzenich et al (2020) revealed a negative relationship between perceived knowledge about large-scale energy infrastructure technologies and associated risk perceptions.When focusing on EFC, electrification is combined with the rather unfamiliar agricultural sector, about which public knowledge is low (Pfeiffer et al 2020).Henceforth, we took perceived knowledge about the sector as the affected system of significance to evaluate the system change toward EFC, even though the cross-sectoral topic leaves the direction of the relationship open.This resulted in the following hypothesis: (H1) Subjective knowledge about present-day agriculture predicts attitudes towards EFC.

Problem awareness
Beyond specific knowledge about a distinct technology or system, general knowledge of energy problems is a crucial precondition for progression of the energy transition (Steg et al 2015).Problem awareness about, for example, environmental problems such as the negative consequences of fossil fuel usage is one influencing factor for environmental behaviors and attitudes in line with value-belief-norm theory (Stern et al 1999).According to value-belief-norm theory, attitudes and problem awareness is determined by individual value orientation and serves as a relevant mediator between value orientation and attitudes (see Van der Werff and Steg 2016).
Research findings have confirmed the importance of awareness of environmental problems for the energy transition by uncovering its relationship to, for example, people's feelings of responsibility for environmental problems, which comes with higher acceptability for energy policies (Steg et al 2005) or a preference for specific energy technologies (Van Rijnsoever and Farla 2014).
Building on the value-belief-norm theory and prior research, we used problem awareness as a proxy for knowledge about the necessity of energy transition and expected its positive impact on the attitudinal components of EFC as a system change: (H2) The general problem awareness of fossil fuels predicts positive attitudes towards EFC.The potential for value-congruence between policies or technologies and individual value orientations in the energy context has been repeatedly shown, even in its early development phase: people judge energy technologies or policies according to specific issues (e.g., protection of the environment) that are important to them (Steg et al 2015, Bouman et al 2021).Therefore, environmental value orientations impact acceptability, risk and benefit perceptions, and favorableness for different energy technologies and sources (De Groot et al 2013, Perlaviciute and Steg 2014).Moreover, Brosch and Sanders (2014) outline the fit between an event and one's core values as a determinant for emotional reactions and decisions.Therefore, it is essential to consider how to design communication strategies for energy system changes that evoke more positive and less negative reactions from the outset.We argue that when people have little knowledge and familiarity with energy system changes, they rely on their basic values as guiding principles.Therefore, we hypothesized that value orientations are distinct predictors of attitudes toward EFC: (H3) Environmental value orientations influence attitudes towards EFC: (H3a) There is a positive relationship between altruistic value orientation and evaluative attitudinal components.
(H3b) There is a positive relationship between biospheric value orientation and evaluative attitudinal components.
(H3c) There is a negative relationship between egoistic value orientation and evaluative attitudinal components.
Additionally, we expected attitude-specific problem awareness to operate as a significant mediator based on the value-belief-norm theory (Stern et al 1999): (H4) Problem awareness mediates the relationship between environmental value orientations and evaluative attitudinal components.

Value-congruent messages
Beyond evaluating attitudes, promoting positive attitudes is key to the acceptance of energy system changes.The question remains of how to actively strengthen positive attitudes at an early stage (Baur et al 2022).At an early diffusion stage for system changes the stability of public attitudes is lower and may change relative to the provision of new information (Bögel et al 2018).Additionally, energy system changes represent innovative ideas that are little known to the public.This opens up the possibility that communication strategies can promote positive attitudes toward system changes, e.g., by informing people about potential benefits.Even providing general information about energy problems can increase knowledge and awareness of environmental issues Thus, value-congruent messaging appears as a promising communication strategy for changes in the wake of energy transitions in different fields.Therefore, we extend Huijts et al 2012) framework with situational cues in the form of value-congruent message framing.Following Nilsson et al (2016), we assume an impact of message framing in the form of value-oriented benefit framing (biospheric, anthropocentric, and combined) on positive attitudes at the early developmental stage of system change: (H5) Benefit framing influences attitudes towards EFC.
In addition, we examined the possibility of targeted group-specific communication by testing the effect of benefit framing that is complementary to one's value orientations.We expected that biospheric framing focusing on environmental benefits would trigger a biospheric value orientation, and anthropocentric framing focusing on benefits to humans would activate egoistic and altruistic value orientations, subsequently resulting in more positive EFC evaluations: (H6) Benefit framing moderates the effect of value orientations on attitudes towards EFC.(H6a) Biospheric benefit framing moderates the relationship between biospheric value orientation and attitudes toward EFC.
(H6b) Anthropocentric benefit framing moderates the relationship between altruistic value orientation and attitudes toward EFC.
(H6c) Anthropocentric benefit framing moderates the relationship between egoistic value orientation and attitudes toward EFC.

Sample and data collection
The data to test our above-mentioned hypotheses were obtained through an open scientific survey panel in Germany in the autumn of 2021.All participants gave written consent to participate in the study and agreed to subsequent research purposes.In total, 349 participants completed the survey; this included all the relevant scales, although seven participants did not provide demographic information.Subsequently, 48 responses had to be removed due to incorrect or missing answers on the attention check item.Additionally, one participant was excluded due to an unreasonable age statement.The remaining sample of 300 participants (57.33% female), ranged from 19 to 84 years (M = 47.65;SD = 15.49), and 60.9% of the sample held an academic degree, which corresponded with the demographics of the panel sample.Over half of the participants (58.6%) were employed in a salaried position or public service job; this was followed by housewives (17.3%), self-employment (10.7%), retirement (11%), and missing data (2.3%).Participants resided in all federal states within Germany.More than half (58%) had no direct connection to the agricultural sector, whereas the remaining had a professional connection and/or friends and family in the sector.On average, participants took 33 min to complete the survey.2

Survey design
To test the value-congruent messaging approach, an experimental vignette methodology was utilized, in which each participant was randomly assigned to one of three conditions (see figure 2, 'Manipulation' cell).Additionally, a control condition was applied to test for an actual framing effect (Palomo-vélez et al 2021).In this type of methodology, intentions, attitudes, and behaviors are assessed by presenting realistic scenarios and manipulating independent variables; the method favors experimental realism and thus has external validity (Aguinis and Bradley 2014).Following Aguinis and Bradley (2014), the same information was provided across all vignettes in the form of a newspaper article.
Due to the unfamiliarity of the average citizen with the agricultural sector, an introduction to the potential system change was provided using a short description and sketch illustrating a farm run by electrical power (figure 3).The sketch was used to make the scenario more tangible and ensure immersion.
Additionally, general information about the EFC system change was used to provide an equal level of knowledge for each participant.To further ensure participants' engagement with EFC and avoid unintentional influence on their opinions, participants were asked to express their first, spontaneous thoughts on the possible use of EFC in the future using an open-ended question.
The experimental condition depicted a newspaper article with different argument foci (anthropocentric, biospheric, combined arguments, and control group) in favor of implementing EFC.Each article was introduced with the same argument: 'This agricultural transition ideally would be communicated from its early development phase to all parties involved and one possibility of doing so would be an article with transregional reach'.Participants were instructed to imagine they would find the presented article in their weekly newspaper.They were asked to read through it in detail and judge its quality.To maintain the storyline, participants were asked to indicate whether they considered the early communication of EFC (via the presented newspaper article, for example) to be a reasonable procedure.
To explore information quality, participants were requested to evaluate the arguments provided in the newspaper article.3Following Nilsson et al (2016) and Huijts et al (2007), opinions on EFC were formed using participants' credibility, informative value, and usefulness evaluations.The assessments were given on a fivepoint Likert scale (1 = 'strongly disagree' to 5 = 'strongly agree').On average, participants tended to agree and rated the arguments as relatively credible, useful, and informative (M = 3.81; SD = 0.87).No significant differences between the evaluations of the three argument conditions were detected (F(1, 215) = 0.05, p = .83).

Measurement of independent and dependent variables
All the items for operationalizing the individual constructs were adapted from previous studies and translated into German through forward and backward translation.The attitudinal constructs-acceptability, risk perception, benefit perception, willingness to pay, and positive and negative emotions-were linguistically adjusted to EFC as the topic of interest.
Table 1 presents the applied scales with the origin of the items used, exemplary item, and scale reliability (Cronbach's alpha; α).All scales had satisfactory to very good reliability.Composite reliability (CR) and average variance extracted (AVE) were calculated; consistent with Fornell and Larcker (1981), most of the factor loadings were above 0.60, had CRs above 0.70, and AVEs above 0.50, indicating good internal consistency and convergent validity for each construct.One exception was egoistic value orientation, which had an AVE of 0.47.Due to its proximity to the 0.50 threshold and an acceptable CR (0.72; Hair et al 2006), we concluded that it had convergent validity.
For all the independent and dependent variables, participants indicated on a seven-point Likert scale how strongly they agreed with each statement.The verbal anchor ranged from 1 = 'strongly disagree' to 7 = 'strongly agree,' except for self-assessed knowledge about present-day agriculture (Pfeiffer et al 2020) which was adapted from Linzenich et al (2021) and used a six-point scale from 1 = 'not informed at all' to 6 = 'very well informed.' For further data on public perceptions of the system change, participants were asked to submit their assessment of whether EFC would realistically occur in Germany in the upcoming 30 years.To account for the possible influence of lack of involvement, participants were asked to indicate how much they felt affected by the transition toward EFC using three response options (not affected, minimally affected, significantly affected).

Data analysis
Data analysis, including preceding correlations, outlier checks, and descriptive statistics, was conducted using R (v4.0.3;R Core Team 2020).Given the high number of variables and a relatively small sample size, we applied a path analysis to test the strength of the antecedences for multicomponent attitudes toward EFC and the valuecongruent information effect using the Lavaan package (Rosseel 2012).In addition to calculating the scale reliabilities, all the variables were examined using confirmatory factor analysis to reduce the risk of measurement error and yielded satisfactory to good fit indices.Thus, the arithmetic means for all the scales were computed for   the path analysis.Due to multivariate normal distribution violation, coefficients were estimated using the maximum likelihood method with robust standard errors and a scaled test statistic following Yuan-Bentler (2000).
Additionally, a bootstrapping procedure following Bollen and Stine (1992) was applied to compute the confidence intervals to quantify uncertainty.

Results
Initially, we tested for multicollinearity between the independent variables, which was acceptable at < 1.8 (Johnston et al 2018).Additionally, we scanned for problematic outliers using olsrr R (Hebbali 2020).Five outliers were detected as being problematic (grand outliers with high leverage).Deleting these and retesting the model led to no considerable differences in the results, hence we kept them in our dataset.Overall, the descriptive results for the attitudinal components yielded a positive evaluation of EFC (table 2).On average, participants found EFC acceptable and of perceived benefit, whereas low-level negative emotions and relatively low-risk perceptions were reported.The individual level of affectedness by the specific system change was also striking: only 21% felt significantly affected, while the majority felt slightly affected or not affected at all (79%).
The correlation matrix that formed the basis for the path analysis generally supported the assumed relationships (table 2).However, of note is how strongly the attitudinal components are related to each other.To ensure the distinctiveness of each construct, we compared different factor structures for the attitudinal component associations using confirmatory factor analysis; according to the fit indices, the six-factor structure proved to be the most appropriate.

Antecedences for attitudinal components and value-congruent information
In the following, we present our results in three steps to test our hypotheses.In the first model, value orientations, socio-demographic variables (age and gender), and the effect of manipulation were tested for their direct effect on the different attitudes (table 3).The model was only just identified and therefore no fit indices can be reported.Gender was not significantly related to any of the attitudinal components, but age was.Thus, gender was not included in the further analyses, whereas age was kept as a covariate.
Overall, Model 1 revealed a significant association between value orientation and the cognitive, behavioral, and affective attitudinal components of EFC (H3).Generally, the stronger the altruistic and biospheric value orientation, the more positive were participants' EFC evaluations.Interestingly, the predicted paths for egoistic, altruistic, and biospheric value orientaiton were only partly significant for the different attitudinal components.Biospheric value orientation only had a significant effect on positive attitudes, whereas altruistic value orientation showed no significant influence on acceptability and affective response.In contrast to our hypothesis, a stronger egoistic value orientation did not correspond with a more negative evaluation.Instead, egoistic value orientation significantly predicted positive emotions and benefit perceptions.Accordingly, we can only partly confirm H3a and H3b, and reject H3c.Moreover, contrary to expectations, the manipulation of the presented arguments did not significantly influence the evaluation of EFC, except for a small effect on risk perceptions in the combined argument foci group compared to the other groups (β = −0.14, p = .042).Based on the confidence interval, we assumed no impact on the evaluation by the type of benefits presented.Consequently, H5 was not confirmed.
In a second model, we added knowledge as a covariate and problem awareness as a mediator (table 3).The robust results of the path analysis indicate that the model had excellent fit according to Hu and Bentler's (1999) recommendations (χ 2 [4] = 5.116, p = 0.276; CFI = 0.99; TLI =.99; RMSEA = 0.029; SRMR = 0.012).Adding both variables as antecedents noticeably increased the variance explained for each attitude component.
As displayed in figure 4, awareness of the problematic nature of fossil fuel turns out to be a significant predictor for all the attitudinal components and confirmed H2.On the contrary, knowledge of the agricultural sector showed a rather small but significant effect on both negative components (negative emotions and risk perceptions).The effect was relatively small and limited to specific attitudinal components, therefore H1 was partly supported.In Model 2, the negative relationship between age and attitudinal components was only assumed for acceptability, positive emotions and problem awareness based on the confidence interval.
Problem awareness proved to be a relevant mediator for two value orientations (H4).The direct effect of altruistic value orientation on the attitudinal components and the remaining direct effect of biospheric value orientation both decreased.The indirect effects, and thus the partial mediation for both value orientations on nearly all attitudinal components, were significant (see appendix table A1 in the supplementary material for the indirect, direct, and total effect paths).However, full mediation only applied to the altruistic value orientation effect on all components and not to the pathways between biospheric value orientation and attitudes, as a direct effect remained.However, for egoistic value orientation on the attitudinal components, there was no mediation effect through problem awareness.Therefore, we can only partly confirm H4 with a limitation to biospheric and altruistic value orientations.
To test the predicted interaction effect between value orientations and corresponding benefit information (H6), three interaction terms were created by multiplying the mean-centered value orientation and the dummy variable.The model with exact effect sizes and p-values can be found in appendix table A2 in the supplementary material.Benefit framing, in contrast to the other conditions, did not significantly moderate the relationship between biospheric value orientation (H6a).Therefore, the relationship between biospheric value orientation and attitudes was not noticeably stronger when biospheric arguments were presented.The same applied to the predicted moderation of the anthropocentric framing, where we found no significant influence on the relationship between egoistic value orientation (H6b) and altruistic value orientation (H6c) and attitudes.Based on the present results, we rejected H6 and concluded that value-congruent messaging did not significantly impact the evaluation of EFC at the early development stage.

Explorative analysis
As we found no effect of different framing, we added the degree of personal affectedness as a covariate.In line with previous research (Schuitema et al 2020, Homar and Cvelbar 2021), individual feelings of affectedness (personal involvement) are presumed to affect the processing of information.Therefore, the high level of unaffectedness in the present sample might be a reason for the missing effect.Indeed, participants with a high level of affectedness perceived more risks (β = 0.18, p > .01)and indicate stronger negative emotions (β = 0.28, p < .00)toward EFC.

Discussion
This study served to obtain a comprehensive understanding of the evaluative response toward EFC, to identify influential factors, and to test a potential reinforcement strategy for positive attitudes.The multidimensional evaluation of EFC as an unknown system change revealed a positive cognitive evaluation and willingness to pay for the realization, while it elicited weak negative and positive emotions.Although the positive finding indicates later acceptance and support, a lack of emotion can have various causes, such as nil to little threat perception or individual significance (Brosch and Steg 2021).This supports the low levels of personal affectedness seen in the present sample.Generally, this points to the distance between the agricultural sector and the public, which was the starting point for this investigation, and contains implications for the interpretation of the results.
In line with previous research (Steg et al 2011, De Groot et al 2013), value orientations were a guiding principle for evaluating EFC as a novel system change.Specifically, biospheric value orientation was related to more positive evaluations.Contrary to our expectations, egoistic value orientation was positively correlated to cognitive, affective, and behavioral attitudes.We suspect that EFC system changes in the wake of the energy transition imply individual benefits and therefore are not congruent with egoistic value orientation (Liu et al 2021).Furthermore, individuals may not primarily perceive individual disadvantages (e.g., time and money) in system change in an unfamiliar sector, which is supported by the low observed emotion and affectedness levels.Interestingly, we found that problem awareness about fossil fuels served as a universal predictor for all attitudinal components, whereas the influence of knowledge about the respective sector was only negatively related to negative attitudinal components.Although environmental value orientation proved to have predictive power for the attitudinal components, the impact of altruistic and biospheric value orientations was partially to fully mediated by problem awareness, which corresponds to findings by, for example, Gifford and Nilsson (2014), who found that norms serve as a full mediator between acceptance and environmental values.Therefore, problem awareness about fossil fuels appears as a relevant prerequisite for different energy system changes, which is in line with the value-belief-norm theory (Stern et al 1999) and emphasizes the importance of continuously transparent communication about the necessity of replacing fossil fuels in multiple industries in order to generate support and acceptance for energy transitions.
Increasing subjective knowledge about the sector was revealed as potentially acceptance-hindering, given the importance of emotions for beliefs and behaviors (Spampatti et al 2022), although its significant influence was limited to negative components and did not notably correlate with lower acceptability.Subjective knowledge as a critical lens on system changes was found by Linzenich et al (2020) regarding large-scale energy infrastructure technologies.Given the parallels between such technologies and large-scale EFC system changes, presumed knowledge about the system seems to lead to increasing concerns about the difficulties, and thus the risks, of system change.This finding indicates the need for more comprehensive communication involving, for example, imparting risk prevention intentions to the public to increase their knowledge and feelings of personal affectedness.
Considering the early diffusion stage of EFC, we expected additional information to influence its evaluation.We hypothesized that value-congruent information would lead to more positive evaluations of EFC compared to non-congruent information.Both hypotheses were not supported, however.Surprisingly, neither a significant main effect of benefit framing nor an interaction of value-congruent messaging emerged.As mentioned above, this might be caused by a lack of affectedness.The current distance between the public and agriculture might be the reason for limited awareness of their own involvement in system change.As raised in the explorative analysis section, a possible lack of concern and interest does impact how attentive people are.The higher the personal interest and affectedness, the more strongly the information is processed, and the more a subsequent change of attitude towards a transition or technology is likely (Schuitema et al 2020).This aligns with former theoretical and empirical research in which personal involvement was found to be a crucial moderator of information framing (Pleger et al 2018, Homar andCvelbar 2021).Building on this, we assumed some degree of information processing in our study.There could also have been an unintended effect of introducing EFC within the energy transition, which generally receives support and is accepted (Andersen et al 2023).The simple localization of EFC as an unfamiliar system change in the energy transition might be a sign of complexity reduction for the participants, but we could not account for this with certainty in our study design.
Taken together, the results have important theoretical and practical implications.They provide relevant insight into the relevance of communicating upcoming energy system changes in specific sectors to the public.The strong correlation between problem awareness and favorable evaluation of EFC does highlight the importance of generally raising awareness about the necessity to reject fossil fuels in all sectors and potentially localize measures and technologies in the energy transition.At the same time, the tendency towards a more critical assessment of system changes by people with more knowledge of the sector and higher perceptions of affect point to the need to develop targeted group-specific communications.People have different requirements for participation that should be taken into account by, for example, including them in decision-making or public discussions (Steg et al 2015).Moreover, the above-mentioned significance of affectedness for processing information regarding specific system changes raises the question of how far people's personal feelings of issue involvement should be mobilized through, for example, raising awareness on the agricultural sector and its significance for climate change mitigation.This could be achieved by offering people the opportunity to experience the agricultural sector and engage with its actors.To do so, formal educational settings such as in schools or universities (e.g.Eck et al 2021) or self-guided farm visits (Ventura et al 2016) provide opportunities for more interactive engagement with the topic, ultimately enhancing topic interest and understanding (Swim et al 2018).This could be especially valuable for individuals who are more critical, allowing them to experience processes and technologies and fostering a two-way dialogue between the public and sectoral actors (see Pfeiffer et al 2020).Furthermore, the lack of a framing effect and the high percentage of people feeling less to not at all affected leaves the causal direction of affectedness and information processing open to question.It would be interesting to pinpoint to what extent the provision of information about future system changes and possible consequences can trigger personal feelings of being affected by system change, or whether such feelings are a necessary condition for information processing.To stimulate involvement and interest in lesser-known sectors such as agriculture, vignette studies like this one or new concepts (e.g., knowledge-based intervention; Geiger et al 2017) might be a valuable starting point.
On a theoretical note, this initial assessment of a public opinion using cognitive, affective, and behavioral components enables us to shed light on the potential reasons for acceptability or lack of acceptability.As investigated here, future communication strategies and necessary measures to address upcoming system change transitions can be identified early on.Individual attitudinal components (e.g., emotions or benefit perception) can therefore be focused in a way to strengthen subsequent acceptance.
Lastly, we want to encourage the application of methods such as vignette designs with a control group to promote experimental realism and internal validity (Ejelöv and Nilsson 2020).
This study has some limitations that must be taken into account.First, the potential implementation of EFC depicted a rather vague scenario as it combined different technologies and processes.This might have influenced the evaluation in different ways, as single aspects of the entire scenario could have evoked unintended positive or negative associations that affected the evaluation, which we did not control for.The complexity level was higher with an increasing quantity of information, which can cause more negative reactions to the scene due to growing skepticism (Braun et al 2018).Therefore, the information should be carefully selected and empirically tested for unknown energy system changes.Thus, we recommend aiming for a more precise presentation of emerging system changes using more detailed methods, such as videos, instead of pictures and text, to avoid selective perceptions.
Second, and with a focus on attitudes as a construct, in an attempt to enhance positive attitudes, we did not record prior attitudes, which can influence reactions toward transition (Bögel and Upham 2018).For future studies on emerging system changes, we suggest including a baseline measure in the form of, for example, a preand post-questionnaire, or taking a longitudinal approach in line with Spampatti and colleagues (2022).
Third, regarding the sample, it has to be noted that distribution via panel limits generalization due to the selfselection of participants, as we can assume they are curious about the topic and interested in the research results.The sample also had one non-negligible feature, as discussed earlier: a low level of affectedness.On the one hand, we suspect the missing manipulation effect to be a feature of the present sample due to the absence of involvement.On the other hand, it might be a phenomenon of the relationship between the public and agriculture, which indicates the necessity of establishing concern and affectedness in the public to successfully make use of communication strategies.This being said, this study lacked a cross-cultural comparison, which does not allow for a generalization across countries, where the acceptance of energy technologies may vary due to, for example, different political contexts (Painuly 2001).
Lastly, our results point to the existence of different target groups in the public.To pinpoint and further investigate attitudes toward system changes, future studies might apply latent class analysis, which is of great value for predicting future behavior (Rhead et al 2018).

Conclusion
To conclude, our study contributes to advancing the understanding of public perceptions of the early developmental stage of a potential system change in the agricultural sector.This allows the anticipation of future reactions and counteract its potential failure due to a lack of public acceptance.Next to value orientations and knowledge about the sector, problem awareness about fossil fuels was found to be a strong predictor for attitudes toward EFC that exemplify energy system changes in agriculture.Furthermore, problem awareness serves as a mediator for the impact of biospheric and altruistic value orientations.Overall, this marks the significance of generally strengthening people's problem awareness to achieve public acceptance and include the public in the early developmental phase of system changes.Beyond this, we could not demonstrate the effect of different information presented in the form of varying types of benefits of future system change, which might be related to the absence of personal affectedness.This points to the importance of affectedness as a key adjusting variable and advocates for further research to obtain information on public attitudes, information processing, and the role of personal affectedness toward system changes in the sector.To raise awareness and foster a sense of personal affectedness, governmental and industrial institutions should invest in designing comprehensive communication strategies when implementing system changes early-on.These strategies should encompass one-way information dissemination, as well as two-way communication to strengthen topic involvement and understanding by the public.Given the necessity of transforming multiple industrial sectors, we hope this study will encourage others to focus on the public perceptions of different emerging system changes; focusing on the changes before they are implemented will allow public perceptions to be integrated at the early development stage.Lastly, we could show the value of employing psychological theories and concepts, especially in the agricultural sector.This sector plays a central role in climate change mitigation and undergoes wide-ranging changes, in which the psychological discipline offers insight to support these changes with a focus on the humancentered perspective.
Appendix.Detailed results on mediation effect of value orientation and value-congruent information

Figure 1 .
Figure 1.Postulated model of antecedences for the multidimensional evaluation of EFC by the public.Note.* H4 represents a mediation effect through problem awareness.
Steg et al 2015), and how information is presented affects the public response (Pleger et al 2018, Homar and Cvelbar 2021, Judge et al 2021).Indeed, situational cues (such as messages) can induce the activation of one's own values (Steg et al 2014), and various studies have verified the use of 'value-tailored' (Judge et al 2021) or 'value-congruent' messages (Borgstede and Andersson 2014, Nilsson et al 2016, Palomo-vélez et al 2021) for different environmental policies and technologies.Different studies with varying research objectives reveal a moderating effect of biospheric value orientation on emotional reactions (Judge et al 2021) and egoistic values on acceptability, next to a main effect of benefit framing (Nilsson et al 2016).

Figure 3 .
Figure 3. Sketch of EFC as a future energy system change in agriculture.

Figure 4 .
Figure 4. Direct effects of antecedents on attitudinal components.Note.Only standardized values of significant paths are displayed; For a better visualization, knowledge is not displayed, but the path coefficients are listed in table 3. * p < .05,** p < .01,*** p < .001.
(Perlaviciute and Steg 2014, Nilsson et al 2016-belief-norm theory, awareness of environmental problems is determined by people's values(Steg et al 2005, Ejelöv andNilsson 2020).Considering individual traits beyond individual predispositions, value orientations are prominent antecedents for environmental attitudes and behaviors(Perlaviciute and Steg 2014, Nilsson et al 2016, Bouman et al 2021).Values generally serve as guiding principles in people's lives, are stable over time, and have considerable impact on people's actions, cognitions, and information processing (Schwartz 1992), which makes them essential influencing factors when addressing attitudes toward novel innovations and prospective changes.Three types of values are particularly relevant in the environmental context: egoistic, biospheric, and altruistic.While altruistic and biospheric value orientations show a positive relationship to pro-environmental beliefs, policy acceptability, and intentions, egoistic value orientation demonstrates a negative relationship with these constructs (Steg et al 2011).

Table 1 .
Operationalization for each dependent and independent variable.
* ) Types of E-PVQ values (Bouman et al 2018) 'It is important to [him/her] to be influential.*)'It is important to [him/her] that every person has equal opportunities.*Items deleted due to factor loadings above 0.5 and unacceptable reduction of AVE below the threshold.VO = value orientation.

Table 2 .
Correlation matrix for the independent and dependent variables.

Table 3 .
Path analysis and standardized beta coefficients for the direct effects with 95% confidence intervals for both models.Coefficients are standardized beta estimates for the direct effects.a Main effects of argument foci were tested by coding three dummy variables with the control group as reference.b Women as reference group.N indicates the number of individuals included in the analysis.* p < .05,** p < .01,*** p < .001.VO = value orientation.