Accuracy-directed climate reasoning: how self-efficacy, interest, confidence and judgments relate to knowledge and reasoning outcomes

Abstract

This study examined how cognitive and motivational factors relate to individuals’ reasoning about climate change and sustainability. Specifically, we investigated whether self-efficacy, interest, and confidence in accomplishing environmental sustainability tasks are associated with plausibility judgments, climate change knowledge, and reasoning directionality. A sample of 503 U.S. adults completed measures assessing these constructs. Structural equation modeling showed that interest, self-efficacy, and confidence were positively associated with plausibility judgments, which in turn related to higher levels of climate change knowledge and accuracy-directed reasoning. Participants demonstrating accuracy-directed reasoning exhibited enhanced self-efficacy, greater interest, higher knowledge, and stronger plausibility judgments compared to those engaging in desired outcome-directed reasoning. These findings suggest that self-efficacy, interest, and critical evaluation skills are meaningfully related with epistemic engagement and scientifically grounded reasoning, and may be relevant targets for future efforts at supporting sustainability-oriented action and decision-making.

1 Introduction

Climate change (CC) is an urgent crisis facing humanity today, posing a global challenge with widespread impacts on ecosystems, economies, and human health. Scientific consensus indicates that human activities, particularly greenhouse gas emissions such as carbon dioxide and methane, are the main drivers of global warming (Ripple et al., 2024). Anthropogenic CC has contributed to more frequent and intense extreme weather events, rising sea levels, and significant biodiversity loss (IPCC, 2023). IPCC (2023) emphasizes the need for urgent, coordinated action, advocating comprehensive strategies including both emission reductions and adaptation measures to protect planetary health.

CC is also a scientific issue of social relevance, situated at the intersection of science, policy, ethics, and society, requiring interdisciplinary approaches to address its challenges. The complexity of CC requires not only scientific understanding, but also public participation, science communication, value-based argumentation and collaborative decision-making to develop sustainable solutions (Fawzy et al., 2020; Zeidler and Nichols, 2009; Zeidler and Sadler, 2023). In this regard, an effective response to CC depends on a well-informed public that not only understands the complexity of the issue, but also feels capable of understanding and reasoning about sustainability-related challenges. In this context, knowledge about CC plays a fundamental role, by equipping individuals with essential information to understand its causes and consequences.

However, knowledge alone may not be sufficient for individuals to evaluate scientific information; they must also grasp the plausibility of scientific claims (e.g., explanations, theories, claims, etc.) (Dole and Sinatra, 1998; Sinatra and Lombardi, 2020). Plausibility judgments play an important role in the construction (and reconstruction) of scientific knowledge and development of scientific literacy (Duit and Treagust, 2003; Lombardi et al., 2016, 2024; Posner et al., 1982). In this study, we focus on plausibility as the judgment about the truthfulness of a claim in the presence of alternative claims (Lombardi et al., 2024) which may help explain how motivational factors connect to climate knowledge and reasoning. In the context of the post-truth era, where misinformation and disinformation about science are widespread, developing the public’s ability to evaluate the plausibility of scientific information is essential for civic and scientific engagement (Dawson and Carson, 2020; Sinatra and Lombardi, 2020; Zummo et al., 2021). Because the cognitive process of plausibility of scientific knowledge encourages evidence-based reasoning and informed decision-making in contexts where individuals are faced with conflicting claims and may strengthen the likelihood of cognitive and emotional engagement (Sinatra et al., 2014; Sinatra and Lombardi, 2020).

Evaluating the plausibility of scientific claims frequently necessitates sustained effort and persistence, particularly when individuals are confronted with complex or conflicting information. Self-efficacy and confidence reflect beliefs about one’s capability to successfully carry out such cognitively demanding tasks and to persist in the face of difficulty (Bandura, 1997; Dole and Sinatra, 1998). In this context, self-efficacy and confidence may impact individuals’ motivation to learn and their willingness to restructure knowledge.

Similarly, topic interest is also an important factor that affects motivation to learn and the use of acquired knowledge in sense-making (Dole and Sinatra, 1998). Dole and Sinatra (1998) suggest that individuals are more inclined to engage deeply with topics that align with their personal interests, which in turn leads to improved comprehension and retention. Furthermore, interest enhances engagement by increasing motivation to seek knowledge and deepening personal relevance (Hidi and Renninger, 2006). These motivational aspects are particularly relevant in informal learning environments and outreach settings, where learners construct meaning through participation (Boyd et al., 2021),

Therefore, examining the extent to which self-efficacy, interest and confidence interact with plausibility judgments, knowledge, and reasoning is important because it may clarify why individuals differ in how they make sense of scientific issues such as CC. This approach not only informs formal education but also supports public science engagement by identifying key factors that shape how people evaluate and reason about scientific information (Chinn et al., 2011; Zeidler and Sadler, 2023).

In addition to individual-level variables, engagement with scientific information is also shaped by broader science communication contexts. Strategic public engagement, defined through audience-specific goals and behavioral outcomes (Besley and Downs, 2025), has emerged as a central concern for ensuring that science reaches and resonates with diverse publics. In this context, motivational dispositions such as self-efficacy and interest interact not only with epistemic cognition but also with the framing and communication of climate-related information. Therefore, it is imperative to comprehend the motivational variables inherent within the overarching science communication systems in order to facilitate effective, trust-based engagement with CC.

Accordingly, the present study examined how cognitive and motivational factors relate to individuals’ knowledge and reasoning directionality about CC. Specifically, we approached self-efficacy, interest, and confidence as interconnected dimensions that may be related with how individuals evaluate scientific claims and reason about climate-related information. Considering these factors together provides a more holistic perspective on how cognitive and motivational processes relate engagement with scientific topics of social relevance. By integrating these constructs within a single framework, the study aims to capture the complex interplay of knowledge and motivations which is relevant for informed decision-making in such contexts. This interaction aligns with recent efforts to support informed public reasoning and critical engagement with science in socially complex contexts (Choung et al., 2020; Goldstein et al., 2020; Zummo et al., 2021). Specifically, our goal was to investigate whether participants’ self-efficacy, interest, and confidence in accomplishing sustainability-related tasks related to their plausibility judgments, knowledge, and reasoning directionality about CC and the environment. The aim is also to examine whether these variables differ based on participants’ reasoning directionality; specifically, whether reasoning is directed toward accuracy or toward a desired outcome.

2 Theoretical framework

2.1 Self-efficacy

Self-efficacy, a core concept in social cognitive theory, refers to individuals’ beliefs in their capacity to execute behaviors necessary to achieve specific outcomes (Bandura, 1997). Dole and Sinatra (1998) state that self-efficacy mediates between learner and content in knowledge construction.

High self-efficacy enables individuals to pursue challenging goals, exert effort, and remain resilient through difficulty (Bandura, 1997; Martinie and Shankland, 2024), fostering a solution-focused approach. Its development is based on four sources, mastery experiences, vicarious experiences, verbal persuasion, and physiological/affective states (Bandura, 1997). Thus, self-efficacy relates not only to individual success but also to individuals’ perceived capacity to engage with complex issues (Zhang and Cao, 2025).

It also plays a critical role in engaging with societal problems, as individuals may avoid engagement without belief in the value of their contributions. Individuals with strong self-efficacy approach difficult tasks with persistence, supporting academic success and sustained motivation (Azila-Gbettor et al., 2021; Chen et al., 2021; Honicke and Broadbent, 2016; Meng and Zhang, 2023; Pajares, 2002; Zysberg and Schwabsky, 2021; Wang and Zhang, 2024).

In the context of CC, self-efficacy has been associated with individuals’ engagement with climate-related issues (Hamann and Reese, 2020). Environmental studies show that self-efficacy motivates individuals’ beliefs in their capacity to contribute meaningfully to sustainability-related efforts by enhancing belief in ability to act sustainably (Jugert et al., 2016; Lauren et al., 2016; Reese and Junge, 2017; Tabernero and Hernández, 2011). Greater self-efficacy, tied to personal and collective capacity, has been linked to climate-friendly orientations (Jankowski et al., 2024; Nelson et al., 2022), and to engagement with more challenging sustainability-related demands (Lauren et al., 2016). We included self-efficacy in our study to reveal its role in linking cognitive and motivational components that shape individuals’ approach to climate-related challenges, underscoring the relevance of these attributes for proactive, solution-oriented reasoning.

2.2 Interest

The inclusion of interest in this study as a psychological state characterized by increased attention and participation. Interest plays an important role as a motivational factor that is effective in learning and knowledge construction (Hidi and Renninger, 2006; Linnenbrink and Pintrich, 2002; Mason et al., 2008). Interest promotes deeper cognitive engagement, enhancing learners’ motivation to process new information and restructure prior knowledge (Dole and Sinatra, 1998; Renninger and Hidi, 2022; Renninger et al., 2023).

Higher levels of interest in addressing complex issues such as CC relate to more scientific knowledge revision. Such relevance can foster deeper processing, promote critical evaluation of conflicting information, and support sustained effort when evaluating complex claims. When individuals find climate-related topics personally relevant, they are more likely to invest effort in understanding and evaluating scientific information (Andre and Windschitl, 2003; Mason et al., 2008; Wang et al., 2022).

Interest itself may emerge as a distinct positive emotional experience during interactions with natural environments, fostering curiosity, exploration, and active involvement (Wang et al., 2025). In this sense, interest is not only a precursor to motivation but also a direct emotional response shaped by environmental affordances and affective appraisals. Activating pro-environmental identities can increase the personal relevance of sustainability topics, thereby fostering interest and sustained attention to climate-related content (Capasso et al., 2025; Wallis and Loy, 2021). Therefore, interest remains a key construct for understanding how individuals process and reason about CC and sustainability issues.

2.3 Confidence

Confidence reflects a general perception that usually occurs after an epistemic act or judgment and refers to the degree of certainty regarding performance or perception (Bandura, 1997; Cramer et al., 2009). It plays a crucial role in knowledge construction, engagement, and scientific evaluation, shaping how individuals interpret and integrate new information (Said et al., 2023). In contrast to self-efficacy, which emphasizes beliefs about one’s capability to perform tasks, confidence reflects a retrospective feeling of certainty about one’s knowledge or decisions (Bandura, 1997). This distinction may be particularly relevant in socioscientific contexts, where individuals may feel confident in their knowledge without engaging in deep epistemic evaluation. Recent studies show that an illusion of knowledge may lead to inflated confidence, limiting individuals’ critical examination of climate-related information (Trémolière and Djeriouat, 2021). Furthermore, cognitive sophistication does not always lead to objective reasoning; instead, it can reinforce biases and prior beliefs rather than promote scientifically accurate evaluation (Kahan, 2013; Kahan et al., 2017; Stanovich et al., 2013).

In the context of CC, reducing false confidence while enhancing trust in scientific reasoning can improve individuals’ ability to critically assess the connection between lines of evidence and competing explanations (Sinatra and Lombardi, 2020). This dynamic interplay between confidence and reasoning directly facilitates knowledge (re) construction, particularly in socio-scientific contexts where uncertainty is inherent (Lombardi et al., 2013; Zeidler et al., 2019). Notably, individual beliefs about personal responsibility and agency also influence how climate information is interpreted and evaluated (Syropoulos and Markowitz, 2022).

Given that reasoning ability and bias are not always aligned, examining confidence in accomplishing sustainability-related tasks offers valuable insight into how individuals engage with scientific knowledge and navigate sustainability-related judgments. Knowledge construction involves not just facts, but also argumentation, perspective-taking, and decision-making under uncertainty (Lombardi et al., 2016; Zeidler et al., 2019). Recent evidence further suggests that climate-related anxiety, when accompanied by trust in science, can serve as a motivational force, mobilizing individuals toward pro-environmental reasoning (Ogunbode et al., 2022).

2.4 Plausibility

Understanding new information often requires individuals to reassess existing beliefs and whether alternative concepts make sense and offer meaningful explanations. Plausibility plays a key role in knowledge revision, as it determines how individuals evaluate and integrate new information and assess the credibility of information sources (Dole and Sinatra, 1998; Lombardi et al., 2024). Posner et al. (1982) indicated that individuals must experience dissatisfaction with their existing concepts and that the new concept must be intelligible, plausible, and fruitful for knowledge (re)construction to occur. Individuals are inclined to preserve their existing knowledge and beliefs, which may render them resistant to constructing knowledge aligned with scientific consensus (Lombardi et al., 2024).

In the post-truth era, resistance may amplify when people consider misinformation and non-scientific explanations to be plausible based on cognitive biases (Sinatra and Lombardi, 2020). The prevalence of misinformation and cognitive biases in this era makes it challenging to question the plausibility of scientific claims in a reasoned manner, emphasizing the need for critical thinking skills to evaluate scientific evidence and reappraise both claims and their sources. In the context of complex scientific topics, individuals learn how interconnected the sources of information are by evaluating the plausibility of claims that may be contradictory and controversial to better assimilate information about the climate crisis and develop alternative explanations (Lombardi et al., 2024). By fostering critical thinking skills and promoting reappraisal of initial judgments, society can better address misinformation and encourage informed engagement with the climate crisis (Lombardi et al., 2016; Sinatra and Lombardi, 2020).

In this study, we conceptualized plausibility as the judgment about the truthfulness of a claim in light of alternative claims (Lombardi et al., 2024). Accordingly, plausibility judgments play a crucial role in terms of triggering and shaping conceptual understanding of complex phenomena, such as CC, by serving as a mental filter that influences the acceptance of new information (Lombardi et al., 2016).

2.5 Knowledge

For decades, educators and researchers have explored how learners construct knowledge, particularly in science learning. A key issue is how individuals revise their knowledge and develop deeper, evidence-based understandings. Knowledge construction is a dynamic process that involves integrating new information with existing conceptual frameworks, often requiring critical evaluation and restructuring of prior knowledge (Dole and Sinatra, 1998; Posner et al., 1982). This process is especially relevant in scientific topics of social relevance, where learners often are engaged in reflective thinking and evaluating competing explanations (Sinatra and Lombardi, 2020).

A fundamental aspect is the interaction between cognitive and motivational factors. Prior knowledge provides a foundation but can also act as a barrier when misunderstandings exist (Lombardi et al., 2024). Encountering new information that contradicts pre-existing beliefs, may lead to cognitive conflict, potentially reinforcing prior beliefs depending on how individuals process the new information (Sinatra, 2005). Learners may be more willing to revise their understanding when they are motivated and willing to critically assess new claims (Lombardi et al., 2016). Plausibility, learners’ evaluation of the truthfulness of scientific claims, play a central role in this evaluative process (Lombardi et al., 2024).

In the context of CC and sustainability, knowledge construction is particularly challenging. CC is a complex, interdisciplinary issue that requires scientific literacy, critical thinking, and informed decision-making (IPCC, 2023). Critically evaluating scientific evidence supports individuals’ understanding of CC and their ability to reason about competing explanations (Lombardi et al., 2016). Strengthening evaluation skills equips individuals to engage meaningfully with sustainability-related issues (Lombardi et al., 2024). Ultimately, knowledge construction goes beyond acquiring information; it cultivates evidence-based reasoning and scientific literacy in contexts characterized by uncertainty and complexity like CC.

2.6 Reasoning: accuracy-vs. directed-outcome directed

Reasoning plays a central role in how individuals engage with complex scientific information, such as CC. Learning in these domains requires integrating scientific evidence with social and ethical considerations (Erduran and Dagher, 2014; Zeidler et al., 2019; Zeidler and Sadler, 2023). Yet, reasoning may not always be objective; instead, it may be shaped by motivations, beliefs, identities, and inaccurate knowledge (Sinatra et al., 2014). Researchers commonly discuss this interaction within the broader concept of motivated reasoning, the process influenced by a desire to fulfill epistemic or desired outcome-directed goals (Kunda, 1990; Ditto et al., 2019). While motivated reasoning is used as a theoretical context in this study, the empirical focus of the study is on reasoning directionality (accuracy- vs. desired outcome-directed reasoning).

We conceptualized motivated reasoning as having two facets: correctness and directionality. Correctness reflects a goal to arrive at conclusions that are accurate, valid, and consistent with group-endorsed epistemic standards, such as those employed by the scientific community to evaluate explanatory coherence and plausibility (Lombardi et al., 2016, 2024). Directionality itself can manifest in two ways: directed toward accuracy or toward a desired outcome (Ditto et al., 2019; Pennycook and Rand, 2019). Accuracy-directed reasoning reflects a desire to reach the correct answer but is nonetheless influenced by identity or prior knowledge when cognitive effort is constrained. Desired outcome-directed reasoning, however, aims explicitly to justify preferred conclusions, regardless of their truthfulness, reinforcing biases and group alignment (Stanovich et al., 2013; Kahan, 2013).

This contrast between epistemic and identity goals echoes prior research emphasizing the central role of plausibility judgments in knowledge construction and evaluation. When individuals pursue epistemic goals, they tend to assess the plausibility of competing claims based on coherence and explanatory power, processes that support conceptual development and scientific thinking (Lombardi et al., 2016, 2024; Sinatra et al., 2014). In contrast, when identity goals dominate, plausibility judgments may become distorted, as individuals selectively endorse claims that affirm existing beliefs or group norms, even in the face of contradictory evidence. This tension between openness to new knowledge and the pull of prior commitments can be a critical juncture for interventions aimed at promoting epistemic engagement, particularly in polarized contexts such as CC (Geiger et al., 2020).

2.7 The present study

Despite the extensive scientific evidence pertaining to CC, individuals’ evaluations of climate-related information are frequently influenced by cognitive and motivational factors that shape the interpretation and reasoning process concerning scientific claims (Bandura and Cherry, 2020; Sinatra and Lombardi, 2020). The present study builds on research in science communication and science education, integrating motivational constructs (self-efficacy, interest, and confidence), epistemic judgment (plausibility), knowledge, and reasoning directionality within a single framework. To examine these relations in an integrated manner, we propose two theoretically motivated models illustrating the relations among the constructs: a hypothesized model (Figure 1) and an alternative model (Figure 2). Because the constructs examined in this study are conceptually interconnected, multiple theoretically plausible pathways among these components may exist. Rather than assuming a single directional configuration, the alternative model reflects a different configuration of connections among the same components. Importantly, because the data are cross-sectional, both models are theoretically motivated representations of covarying relations rather than tests of causal ordering.

Figure 1

Figure 2

In line with research that highlights the communicative conditions of epistemic engagement (Wilkinson et al., 2023), we also explore how motivational constructs may be responsive to the way climate information is framed, delivered, and scaffolded in the public sphere. Interest promotes engagement with CC-related content (Renninger et al., 2023), yet its interaction with plausibility judgments and self-efficacy and confidence remains underexplored. Similarly, while self-efficacy and confidence are known to support engagement with complex problems, little is known about how they shape reasoning processes, particularly in the distinction between accuracy-directed and desired outcome-directed reasoning (Kahan et al., 2017; Trémolière and Djeriouat, 2021).

By examining these relations simultaneously, the study aims to clarify how motivational and epistemic factors jointly shape reasoning about CC, rather than examining each construct in isolation. Within these purposes, this study sought to answer the following research questions:

• Does participants’ self-efficacy, interest, and confidence to accomplish environmental sustainability tasks relate to their plausibility judgments, knowledge, and reasoning about climate change and the environment?

• How do these variables differ based on participants’ reasoning about environmental sustainability (directed toward accuracy or directed toward a desired outcome)?

This integrative approach provides the basis for the specific hypotheses outlined below.

2.8 Hypotheses

Based on the model presented in Figure 1, the following hypotheses specify the proposed relations among motivational factors, plausibility judgments, knowledge, and reasoning directionality.

H1: motivational factors and plausibility judgments.

Research on knowledge construction indicates that increased interest promotes deeper learning and acceptance of scientific explanations (Renninger et al., 2023). Likewise, plausibility judgments support the integration of new information (Lombardi et al., 2024). As individuals perceive climate explanations as plausible, they tend to exhibit greater self-efficacy and confidence (Sinatra and Lombardi, 2020). Accordingly, we predicted positive relations between each motivational factor and plausibility judgments. Specifically, we expected:

• H1a: self-efficacy to be positively related to plausibility judgments.

• H1b: interest to be positively related to plausibility judgments.

• H1c: confidence to be positively related to plausibility judgments.

H2: mediating role of plausibility judgments.

According to plausibility-focused models (Lombardi et al., 2016; Sinatra and Lombardi, 2020), individuals are more likely to engage in and adopt scientific ideas they find plausible. Thus, we predicted that plausibility judgments will mediate the relationships between motivational factors (self-efficacy, interest, and confidence) and climate change knowledge and reasoning.

H3: Reasoning: directed toward accuracy or toward a desired outcome.

Research on motivated reasoning distinguishes between accuracy-directed and desired outcome-directed reasoning, which differ in how individuals evaluate scientific evidence (Sinatra et al., 2014; Ditto et al., 2019). Accuracy-directed reasoners evaluate scientific evidence more impartially, whereas desired outcome-directed reasoners often disregard conflicting evidence, dampening plausibility judgments, overall knowledge gains, and confidence in evaluating scientific information (Sinatra et al., 2014; Ditto et al., 2019). Therefore, we predicted that accuracy-directed reasoners will exhibit higher self-efficacy, interest, confidence, plausibility judgments and CC knowledge than desired outcome-directed reasoners.

To our knowledge, this is among the first empirical studies to examine how self-efficacy, interest, confidence, and plausibility relate in a single model of reasoning directionality about CC. By investigating this constellation of motivational and cognitive factors, the study aims to contribute to broader efforts to understand engagement in scientific thinking within real-world contexts of uncertainty and controversy.

3 Methodology

3.1 Research design

We used a cross-sectional, quantitative design to examine structural relations associations among the variables. Additionally, we used a comparative design to investigate how these variables differ based on participants’ motivations for reasoning about environmental sustainability, particularly whether there is a difference between accuracy-directed and desired outcome-directed reasoning.

3.2 Settings and participants

We recruited a sample of 503 individuals residing in the United States, representing a diverse population in terms of gender, age, and race/ethnicity (Table 1). A majority identified as female (55.1%), followed by male (42.5%), while 2.4% identified as non-binary or preferred to self-identity. Ages ranged from 18 to 78, with a median age of 35. In particular, 82% were 50 years old or younger, indicating a relatively young overall. Regarding racial and ethnic background, 62.6% of participants identified as White, and 23.3% as Black or African American. Asian individuals represented 7.2% of the sample, while Hispanic or Latino individuals constituted 3.6%. Smaller proportions identified as belonging to Two or More races (2.0%), American Indian or Alaska Native (0.9%), and Native Hawaiian or Other Pacific Islander (0.4%). This distribution demonstrates the predominance of White participants while reflecting representation from a diverse range of other racial and ethnic groups.

3.3 Instruments and data collection

We employed five instruments, described in detail below, to measure participants’ self-efficacy, interest, confidence, plausibility judgments, CC knowledge and reasoning directionality.

3.3.1 Self-efficacy

We used an adapted version of an existing, previously validated self-efficacy scale to assess participants’ belief in their ability to perform environmentally sustainability-related tasks. We adapted the scale from Bailey et al. (2017) with minor modifications to ensure alignment with the studied topic (original α = 0.94 at pre-instruction and 0.91 at post-instruction). Participants rated each item on a 6-point Likert type scale ranging from 1 (I cannot do this at all) to 6 (I certainly can do this). The scale demonstrated excellent internal consistency in the present sample (ω = 0.93).

3.3.2 Interest

As with the self-efficacy measure, we used an adapted version of an existing, previously validated interest scale to measure participants’ level of interest in topics related to CC. We adapted the scale from Bailey et al. (2017) with minor adjustments to align items with the context of climate change (original α = 0.82 at pre-instruction and 0.87 at post-instruction). Participants responded using a 5-point Likert type scale, where 1 indicated “not at all interested” and 5 indicated “very interested.” The reliability in the present sample indicated excellent internal consistency (ω = 0.92).

3.3.3 Confidence

We assessed confidence as participants’ confidence in accomplishing a range of environmental sustainability-related tasks. The measure consisted of 11 items rated on a 5-point Likert type scale, with 1 indicating “not at all confident” and 5 indicating “very confident.” In the present sample, responses showed excellent internal consistency (ω = 0.94).

3.3.4 Plausibility

In this study, we measured CC plausibility judgments using the Plausibility Perception Measure (PPM) scale, originally developed by Lombardi and Sinatra (2012) and updated by Hanedar et al. (2024). The PPM scale was designed to assess the plausibility judgments about CC. All 15 items in the scale are based on empirically grounded scientific findings from the IPCC (2022) report. It has 15 items, and participants rated the plausibility of the statements on a 10-point Likert type scale, from 1 to 10, indicating 1 being greatly implausible (or even impossible) and 10 being highly plausible. For the analyses, we used an overall plausibility score reflecting participants’ evaluations of the plausibility of climate-related scientific claims. Reliability of the scale represents an excellent consistency measure in the present sample (ω = 0.94).

3.3.5 Knowledge and reasoning

We used the Reasoning about Socioscientific Issues (RASSI) measure to assess CC knowledge and reasoning directionality (Bailey et al., 2025). RASSI is a three-part instrument designed to evaluate (a) content knowledge, (b) reasoning, and (c) confidence in reasoning, about a range of topics (e.g., the causes of CC, its effects on extreme weather events, and issues related to freshwater availability), and has 12 items.

In the present study, only part a and part b of the RASSI were used. Part a consists of items presenting a scientific claim, which participants rate on a 6-point Likert type scale from 1 (“strongly disagree”) to 6 (“strongly agree”), indicating their level of agreement. Part b involves the selection of the most appropriate of the four alternative justifications (reasoning statements) to support their initial rating in Part a. Each item in part b provides four options, along two dimensions to simultaneously assess the epistemic quality of reasoning (correctness) and the motivational stance behind it:

• Correctness of the knowledge: Whether the justification reflects scientifically accurate or inaccurate.

• Type of reasoning: Whether the justification reflects accuracy-directed or desired outcome-directed reasoning.

The categorization of reasoning groups according to combinations of accuracy and motivational orientation is as follows:

• Correct knowledge with accuracy-directed reasoning (CA),

• Correct knowledge with desired outcome-directed reasoning (CD),

• Incorrect knowledge with accuracy-directed reasoning (IA),

• Incorrect knowledge with desired outcome-directed reasoning (ID).

Note that this composite indicator is specific to the design and theoretical assumptions of the RASSI and is not intended to represent a general or fully psychometric measure of reasoning ability.

Part C, which assesses confidence in one’s selected justification, was intentionally not included in the analyses because confidence was measured separately with another scale focused on confidence in accomplishing sustainability-related tasks.

For this study, we retained seven items directly related to CC after removing five items addressing freshwater because they did not align conceptually with the study’s primary focus. Although freshwater issues are environmentally relevant, these items reflected domain-specific knowledge and behaviors that could introduce construct-irrelevant variance and obscure the interpretation of motivational and epistemic responses specific to CC.

The scale demonstrated good internal consistency for Part A (ω = 0.82), while Part B yielded a lower coefficient (ω = 0.68), falling just below the conventional 0.70 threshold often cited in the literature (Nunnally, 1978). While this coefficient is marginally below the suggested threshold, we consider it acceptable for several reasons. First, the scale captures a nuanced construct that may naturally involve a broader range of item responses, leading to modest internal consistency. Second, this level of reliability aligns with findings from our prior research employing Model-Evidence Link instructional tools, where similar reliability coefficients for reasoning-related measures have been deemed adequate given the exploratory nature of the work. Moreover, as supported by past reviews of behavioral research (Peterson, 1994), reliability estimates in the 0.65–0.70 range are frequently accepted when theoretical relevance and interpretability are preserved. Given all this, we retained Part B and considered its internal consistency to be adequate for the analysis.

At the same time, we acknowledge that Nunnally’s (1978) discussion of thresholds was primarily in the context of bivariate analyses, and that the implications of lower reliability in multivariate contexts such as SEM are less certain. Accordingly, we interpret findings involving Part B with caution, and we highlight this limitation in our discussion while noting the need for future research to refine this measure and thereby strengthen its applicability in SEM analyses.

3.4 Data collection

We carried out the procedures online using Prolific, a crowdsourcing platform for participant recruitment, and administered the surveys through Qualtrics, an online research platform used for data collection and analysis. Before administering the questionnaires, we randomized the order of the scales. Since the sequence of scales or questions may influence participants’ responses and introduce order effects or response bias (DeVellis and Thorpe, 2021), we implemented randomization to reduce such biases and enhance the validity and reliability of the data. To ensure consistent and valid measurement, we randomly presented five different scales to participants in varying orders.

We first obtained informed consent from all participants, providing them with a clear explanation of the study’s purpose, procedures, and potential risks. Then participants answered three demographic questions regarding their gender, age, and race/ethnicity. Thereafter, they completed a total of 51 items across five different scales. The data collection process took approximately 20 min. After participants completed the study, we provided a debriefing to explain the aims of the research, clarify any potential misunderstandings, and reduce the risk of emotional reactions beyond the study context.

3.5 Data analysis

We began the data collection process by gathering responses from 508 participants via Qualtrics. After collecting the data, we tabulated and organized the dataset for analysis. We implemented data quality checks to ensure the reliability of participant responses. Specifically, we examined survey completion times and response patterns across scales. Participants who completed the survey in an unusually short time (e.g., under approximately 2–3 min), or who provided highly uniform (e.g., selecting the same response option for all items within a scale, such as consistently endorsing the highest value on the self-efficacy items) or inconsistent responses across items (e.g., alternating between extreme response options across adjacent items), were flagged and reviewed as part of the quality assurance process. These steps enhanced the integrity of the dataset by reducing inattentive or careless responding. During data cleaning, we identified and removed five outliers, resulting in a final sample size of N = 503 participants. We conducted an a priori power analyses using G*Power (Faul et al., 2009) to determine a minimum sample size required to detect medium-sized effects (Cohen’s f = 0.25 for ANOVA; f² = 0.15 for SEM) with an alpha level of 0.05 and power of 0.80. For these analyses and based on our estimated number of groups and parameters, the required sample size was approximately 150–200 participants. Our final sample size of 503 exceeded these thresholds, indicating sufficient power to detect medium effects in both analyses.

We calculated descriptive statistics, including means, standard deviations, and frequencies, for demographic variables (gender, age, and race/ethnicity) and scale responses.

To address the research questions, we applied distinct analytical approaches tailored to each question. For the first research question, which examined the relationships among participants’ interest, plausibility judgments, knowledge, self-efficacy, and confidence, we conducted Pearson correlation analyses to evaluate the strength and direction of associations among the variables. Subsequently, we used structural equation modeling (SEM) to test the hypothesized and alternative models, integrating the correlational findings to examine the structural relationships among variables and evaluate model fit.

For the correlational and structural analyses, reasoning was operationalized as a composite variable derived from participants’ responses on RASSI Part B. Each response reflects a combination of epistemic correctness (correct vs. incorrect) and reasoning directionality (accuracy-directed vs. desired outcome-directed). While many may consider these components as conceptually distinct, our research and development work in developing the RASSI, suggest that individuals’ epistemic accuracy and reasoning orientation often co-occur in dynamic and meaningful ways. Accordingly, rather than isolating these as two constructs, this composite reflects how correctness and reasoning orientation interact in thinking about scientific issues of social relevance, capturing a practical, integrated indicator of reasoning quality aligned with the design of the RASSI instrument.

Responses were coded into four ordered categories: incorrect desired outcome-directed (ID = 1), correct desired outcome-directed (CD = 2), incorrect accuracy-directed (IA = 3), and correct accuracy-directed (CA = 4). This ordering reflects the degree to which responses align with epistemic goals. Although CD responses are scientifically accurate, they were assigned a lower numerical value than CA and IA because they reflect reasoning driven by a desired outcome rather than a commitment to epistemic accuracy. In contrast, IA responses, despite being incorrect, reflect an accuracy-directed reasoning and were therefore coded higher than CD. Higher values indicate greater alignment with accuracy-directed reasoning outcomes.

While reasoning was treated as an ordered composite indicator in correlational and structural analyses, group-based analyses treated the same construct categorically to facilitate interpretable comparisons across distinct reasoning profiles. Accordingly, for the second research question, the four reasoning profiles (CA, CD, IA, ID) were retained as categorical groups in the ANOVA (Analysis of Variance) analyses. This approach facilitated interpretable comparisons of self-efficacy, confidence, interest, plausibility judgments, and knowledge across the four reasoning groups and allowed us to examine how reasoning directionality is associated with these dependent variables. Because the different dependent variables are conceptually different and with some of the dependent variables measured on different scales, we performed separate ANOVAs with Bonferroni-adjusted post hoc tests to support clearer interpretation and understandability of individual effects while limiting Type I error.

4 Results

4.1 Preliminary analyses

We conducted preliminary analyses using JASP Team (2024), an open-source statistical software, to assess the distribution of the data and ascertain the validity of the assumptions underlying subsequent statistical analyses. These analyses included the computation of descriptive statistics for the study variables, such as the mean, standard deviation, standard error, range, skewness, and kurtosis. The mean score for self-efficacy was 50.3 (SD = 8.08), interest was 21.6 (SD = 5.91), confidence was 33.4 (SD = 11.1). Plausibility judgments averaged 114 (SD = 22.3). Knowledge averaged 32.8 (SD = 5.66). All skewness and kurtosis values fall within ±1, which is a satisfactory range for further ordinary least squares (OLS) parametric analyses (West et al., 1995; Table 2). Reasoning profiles were summarized using frequencies and percentages. The majority of participants were classified as showing correct, accuracy-directed reasoning (CA), while fewer participants fell into the CD, IA, and ID reasoning categories (Table 3).

To examine the relationships among the key study variables, we first computed Pearson correlation coefficients (Table 4). The results indicated consistent positive correlations among self-efficacy, interest, confidence, plausibility judgments, and climate change knowledge. Plausibility judgments showed particularly strong correlations with climate change knowledge and moderate correlations with reasoning. Reasoning was more weakly, but significantly, correlated with plausibility judgments, climate change knowledge, self-efficacy, and interest, while its association with confidence was not statistically significant. Overall, the correlational pattern was consistent with the hypothesized relations and informed the subsequent structural equation modeling analyses.

4.2 Structural equation modeling

The preliminary results provided a foundation for our SEM analysis, enabling a more comprehensive examination of the interrelationships among the variables, and highlighting the potential role of plausibility judgments as a mediator in shaping knowledge construction. We used WarpPLS 8.0 to test our hypothesized model, which is a variance-based SEM software that facilitates the testing of complex theoretical models by estimating both linear and nonlinear relationships among latent constructs and provides robust assessments of model fit and predictive quality (Kock, 2017).

4.2.1 Hypothesized model

The results demonstrated strong statistical significance and excellent model fit for the hypothesized model (Figure 3). Model fit indices indicated strong explanatory performance [The Tenenhaus Goodness-of-Fit (GoF) = 0.683] exceeded the recommended threshold of 0.36, indicating a robust explanatory model (Wetzels et al., 2009). The average standardized path coefficient (APC = 0.269, p < 0.001) suggested meaningful relationships among the variables within the hypothesized model (Kock and Lynn, 2012). The average R-squared (ARS = 0.492, p < 0.001) and adjusted average R-squared (AARS = 0.489, p < 0.001) also showed substantial explanatory power, well above the acceptable threshold commonly recommended (Sarstedt et al., 2021).

Figure 3

As shown in Table 5, individual standardized path coefficients revealed several strong and statistically significant relationships. Interest demonstrated a strong direct effect on plausibility, and plausibility had a direct effect on knowledge. In addition, knowledge showed a strong direct effect on reasoning, and plausibility demonstrated direct effects on reasoning, highlighting the pivotal relational role of plausibility judgments with students’ reasoning processes (Lombardi et al., 2024). Confidence in accomplishing sustainability-related tasks showed a small but significant direct effect with plausibility judgments but not on knowledge, indicating a more limited role within the structural model.

Furthermore, indirect effects were substantial; notably, interest and self-efficacy to reasoning, aligning with previous mediation guidelines that emphasize indirect relationships’ theoretical significance (Preacher and Hayes, 2008; Zhao et al., 2010). While plausibility also demonstrated indirect effect on reasoning, self-efficacy and interest showed indirect effects on knowledge.

Effect size (f²) estimates further underscored the importance of key predictors. Interest (f² = 0.264) and self-efficacy (f² = 0.159) demonstrated medium-to-large effects on plausibility, while plausibility itself had a large effect on knowledge (f² = 0.427). Knowledge also showed a large effect on reasoning (f² = 0.348), whereas plausibility’s direct effect on reasoning was smaller but meaningful (f² = 0.110). These results support the hypothesis that plausibility and knowledge play a mediating role in linking motivational constructs and reasoning. Additionally, confidence showed only very small effects on plausibility (f² = 0.024) and knowledge (f² = 0.002).

Reliability and validity indicators further confirmed the quality of the measurement model. Composite reliability values ranged from 0.824 to 1.000, comfortably exceeding the commonly recommended threshold of 0.70, indicating high internal consistency (Sarstedt et al., 2021). Convergent validity was established with AVE values exceeding 0.701, above the threshold of 0.50 (Fornell and Larcker, 1981; Sarstedt et al., 2021). Discriminant validity was confirmed by the Fornell-Larcker criterion, as the square roots of AVEs were higher than inter-construct correlations (Fornell and Larcker, 1981). Full collinearity variance inflation factors (VIFs ≤ 2.858) were also below the recommended threshold (ideally ≤ 3.3), indicating no problematic multicollinearity within the structural model (Kock and Lynn, 2012).

Overall, the SEM explained substantial variance in the key outcome variables, accounting for 45% of variance in plausibility, 57.0% of variance in knowledge, and 46% in reasoning. These results indicate a robust model, demonstrating meaningful relationships and high explanatory capacity consistent with recent structural modeling guidelines and practices (Kock, 2017; Sarstedt et al., 2021; Wasserstein et al., 2019).

4.2.2 Alternative model

The alternative model also demonstrated strong statistical significance and excellent model fit (Figure 4). The GoF value (0.663) exceeded the recommended threshold of 0.36, although it was slightly lower than the hypothesized model. Similarly, APC (0.213), ARS (0.462), and AARS (0.458) values indicated meaningful but comparatively weaker explanatory performance than the hypothesized model.

Figure 4

As shown in Table 5 and Figure 3, standardized path coefficients revealed several strong and statistically significant relationships. Interest demonstrated a direct effect on reasoning, while self-efficacy and confidence also showed significant effects. In turn, there was a direct relational path between plausibility and reasoning, as well as knowledge. Similarly, there was a direct relational path between reasoning and knowledge. Additionally, self-efficacy and confidence also demonstrated direct effects on knowledge. Beyond these direct effects, several indirect effects also emerged. Both self-efficacy and interest showed indirect effects on plausibility, whereas confidence did not. Similarly, self-efficacy, interest, and reasoning demonstrated significant indirect effects on knowledge. Confidence, however, did not demonstrate a significant indirect effect on knowledge.

These effects were robust, yet the overall explained variance in reasoning was notably lower than in the hypothesized model, while the explained variance in knowledge, and in plausibility was higher.

Effect size (f²) estimates further underscored the importance of key predictors. Reasoning had a medium effect on knowledge (f² = 0.234) and plausibility had a large effect on knowledge (f² = 0.314). Interest showed a small-to-medium effect (f² = 0.167), self-efficacy a small effect (f² = 0.056), and confidence a very small effect (f² = 0.015) on reasoning. Additionally, interest (f² = 0.175) and self-efficacy (f² = 0.120) had notable effects on plausibility, whereas confidence had no effect. Compared with the hypothesized model, these effects indicated weaker explanatory power for reasoning but relatively stronger prediction of knowledge.

Reliability and validity indicators further confirmed the quality of the measurement model. Composite reliability values ranged from 0.824 to 1.000, exceeding the recommended threshold of 0.70 (Sarstedt et al., 2021), same as the hypothesized model. Convergent validity was established with AVE values ≥ 0.701, above the threshold of 0.50 (Fornell and Larcker, 1981; Sarstedt et al., 2021). Discriminant validity was confirmed by the Fornell-Larcker criterion (Fornell and Larcker, 1981), and full collinearity variance inflation factors (VIFs ≤ 2.858) were below the recommended threshold (≤ 3.3), consistent with the hypothesized model (Kock and Lynn, 2012). Overall, the SEM explained substantial variance in the key outcome variables, accounting for 49.5% of variance in plausibility, 65.4% in knowledge, and 23.8% in reasoning.

Both the hypothesized and alternative models demonstrated adequate reliability, validity, and overall explanatory power, thus confirming their robustness for understanding the relations among motivational constructs, plausibility, knowledge, and reasoning. Although the alternative model explained slightly more variance in knowledge, this difference was marginal and accompanied by substantially lower explained variance in reasoning outcomes. In contrast, the hypothesized model offered a more parsimonious and theoretically aligned representation of the relations, particularly with respect to reasoning, while the alternative model highlighted additional, but more complex patterns of associations among the same constructs. Importantly, this comparison is descriptive rather than confirmatory; both models represent theoretically plausible configurations of covarying relations, and differences in fit and explained variance are reported to aid interpretation rather than to establish causal or directional superiority.

4.3 ANOVA analysis

To examine how individuals’ reasoning (accuracy-directed vs. desired outcome-directed) relate their self-efficacy, interest, confidence in accomplishing sustainability-related tasks, plausibility judgments, and knowledge, we conducted a one-way ANOVA using four reasoning profiles derived from Part B: Correct knowledge with accuracy-directed reasoning (CA), correct knowledge with desired outcome-directed reasoning (CD), incorrect knowledge with accuracy-directed reasoning (IA), and incorrect knowledge with desired outcome-directed reasoning (ID). Post hoc comparisons used Bonferroni correction, and Welch’s adjustment was applied when homogeneity of variances was violated.

Significant group differences emerged for self-efficacy, interest, plausibility judgments and knowledge, but no significant differences for confidence (Figure 5). Detailed descriptive statistics for each dependent variable are presented in Table 6. Self-efficacy showed small-to-moderate effects [F(3, 499) = 12.026, p < 0.001, η² = 0.067, 95% CI (0.028, 0.110)], with CA scoring highest scores. Interest also differed significantly across reasoning profiles [F(3, 499) = 23.957, p < 0.001, η² = 0.126, 95% CI (0.074, 0.178)], indicating a large effect and exhibiting a similar pattern. Confidence in accomplishing sustainability-related tasks did not significantly differ across reasoning profiles [F(3, 499) = 2.018, p = 0.110, η² = 0.012, 95% CI (0.000, 0.033)], indicating a very small effect size. Plausibility judgments showed the largest effect [F(3, 499) = 45.404, p < 0.001, η² = 0.214, 95% CI (0.153, 0.273)], indicating a large effect size, with the CA group reporting substantially higher scores than all other groups.

Figure 5

We performed a Kruskal-Wallis test to gauge knowledge differences [χ²(3) = 108.02, p < 0.001, ε² = 0.215, large effect size, CI (0.159, 0.285)]. Similar to plausibility scores, the CA group outperformed all others.

The results indicated that reasoning was positively associated with plausibility judgments, knowledge, interest, self-efficacy, but not with confidence in accomplishing sustainability-related tasks. Regardless of reasoning direction or knowledge accuracy, individuals reported similar levels of confidence in performing sustainability-related tasks. Notably, participants in the CA group consistently demonstrated significantly higher scores across cognitive and motivational constructs compared to desired outcome-directed reasoners, suggesting that accuracy-directed reasoning may play a key role in fostering engagement with climate-related issues.

5 Discussion

This study examined how self-efficacy, interest, and confidence relate to plausibility judgments, knowledge construction, and reasoning about climate change (CC). Our hypothesized model positioned plausibility as a statistical mediator linking these factors with knowledge. In this study, greater plausibility refers to participants’ personal judgments about the truthfulness of scientific statements on climate and sustainability drawn from the most recent IPCC report (Hanedar et al., 2024). Results from the structural model showed that interest and self-efficacy had moderate-to-strong positive relations with plausibility, while confidence was more weakly but still positively related. In turn, higher plausibility had strong relations with knowledge. These findings align with Lombardi et al.’s (2024) proto-theory of scientific thinking and earlier frameworks identifying plausibility as a central judgment in science learning (Lombardi et al., 2016). They also echo work suggesting that reflective engagement with relevant scientific issues can foster reappraisal of epistemic judgments (Sinatra et al., 2014). Within this pattern, plausibility appears to emerge as a central linking construct through which self-efficacy, interest, and confidence relate to knowledge outcomes.

Knowledge also emerged as a statistical mediator within the tested models. In our hypothesized model, higher plausibility had strong relations with greater knowledge, and greater knowledge had positive relations with accuracy-driven reasoning. Group comparisons reinforced this pattern. Participants in the CA (correct and accuracy-directed) reasoning group consistently scored higher on self-efficacy, interest, knowledge, and plausibility judgments than those in the desired-outcome groups. Taken together, these results indicate that self-efficacy, interest, and confidence relate to plausibility, plausibility relates to knowledge, and knowledge relates to reasoning. By situating reasoning within the structural model, the hypothesized model describes how self-efficacy, interest, confidence, and plausibility judgments are structurally linked with knowledge and reasoning. This framing avoids claims of temporal causality but underscores the potential for plausibility judgments and knowledge to function as statistical mediators in supporting reasoning that is more consistent with scientific evaluation practices, including the consideration of evidence, the coherence of claims, and the relative truthfulness of competing explanations.

For comparison purposes, we also tested an alternative model to assess whether different structural pathways could account for the observed relations. The alternative model provided a more complex and layered structure that captured multiple indirect pathways; however, this complexity also introduced interpretive challenges and weaker explanatory clarity for reasoning. This complexity also reduces accessibility for instructional practice, which comes at the cost of accessibility, as educators may find it harder to translate into straightforward strategies for practice. In contrast, the hypothesized model offered a more straightforward and parsimonious representation of the relations, making it easier to be more interpretable, applicable, and translatable into practical implications for teaching and learning, and which may also serve as a detailed instructional guide. Moreover, the hypothesized model aligned more closely with theoretical backgrounds regarding the role of plausibility (see 2.4 Plausibility section), offering a clearer and more coherent account of how motivational factors relate plausibility judgments, knowledge construction, and reasoning. When considered as a whole, both models are statistically adequate, however, the hypothesized model offers greater theoretical and practical sense, particularly in terms of highlighting reasoning as a pivotal explanatory mechanism.

Within this regard, as hypothesized, higher interest and self-efficacy were correlated with stronger plausibility judgments, which in turn predicted greater knowledge and accuracy-directed reasoning. SEM analysis indicated that plausibility as a key statistical mediator between motivational constructs and reasoning. While group comparisons showed that individuals demonstrating accuracy-directed reasoning (CA group) scored higher on self-efficacy, interest, knowledge and plausibility, confidence did not differ significantly across reasoning types. Overall, these findings support our proposed model and hypotheses, suggesting that motivation and cognitive evaluation jointly contribute to scientifically grounded reasoning about CC.

Taken together, the results suggest that plausibility judgments were consistently related to self-efficacy, interest, and confidence, and statistically linked with knowledge and reasoning within the tested models. This result is consistent with Lombardi et al.’s (2024) prototheory on development of scientific thinking, which posits that reappraisal of epistemic judgments, such as plausibility, can act as a mechanism for knowledge building and reasoning. Our findings may therefore be understood as aligning with the idea that greater plausibility reflects more than surface-level agreement, potentially indicating greater alignment with explanations that are coherent and personally meaningful (Herrick et al., 2023).

In addition to these motivational and communicative perspectives, recent research on socioscientific reasoning highlights the epistemic underpinnings of how individuals justify scientific claims. Bader et al. (2023) found that undergraduate students often rely on multiple types of justification, such as personal, authoritative, and identity-based, when making decisions on socioscientific issues. This theoretical work complements our interpretation of plausibility as an epistemic judgment that is shaped by context and identity. In particular, plausibility judgments may be indicative of the manner in which individuals reconcile competing claims about a phenomenon when considering evaluations of information sources and their credibility. Future research may explore the manner in which self-efficacy and interest interact with justification-for-knowing patterns to explain variation in people’s epistemic judgments and engagement.

The distinction between accuracy-directed and desired outcome-directed reasoning was also notable. CA group participants, those demonstrating reasoning that was correct and accuracy-directed, scored higher across all constructs. Conversely, ID group participants, those demonstrating reasoning that was incorrect and desired outcome-directed, had lower scores. These findings align with prior research revealing that individuals can direct their reasoning to support prior beliefs, even when faced with conflicting evidence (Kahan et al., 2017; Pennycook and Rand, 2019; Sinatra et al., 2014; Stanovich et al., 2013; Trémolière and Djeriouat, 2021).

5.1 Implications for science communication, engagement, and decision-making

These findings also have direct implications for science communication and public engagement. As Besley and Downs (2025) emphasize, the perceived advantages of structured communication strategies and institutional trust significantly influence public engagement endeavours in the domain of environmental science. The integration of variables, such as interest and self-efficacy, into science outreach initiatives has the potential to enhance public engagement and more accurate reasoning (Teshera-Levye et al., 2025). This improvement can support developing scientific and civic agency through meaningful participation (Bandura and Cherry, 2020). In this context, plausibility judgments can serve as a focal point for both individual reasoning and decision-making in public engagement, linking personal evaluations with broader conversations about climate and sustainability. Consequently, science educators and communicators should consider how message framing, engagement design, and evidence scaffolding interact with self-efficacy and interest to support scientifically grounded, civically engaged, sustainable behaviors (Levy et al., 2021; Smith et al., 2023). However, empirical evidence also highlights how interest and self-efficacy can be shaped and constrained by the broader professional and institutional environments in which science communication and learning occur (Vulturius and Gerger Swartling, 2015; Wilkinson et al., 2023). Therefore integrating contextual insights into models of plausibility judgments and reasoning has the potential to elucidate the processes through which climate supportive and sustainable affect and beliefs are either activated or inhibited in public engagement and learning settings.

Our findings also underscore the complexity of reasoning, which is influenced by both epistemic cognition and beliefs (Chinn et al., 2011; Kahan et al., 2017). Our model suggests meaningful relations in which motivational factors such as self-efficacy and interest predict plausibility evaluations, which in turn, predict scientific knowledge and facilitate reasoning. This pattern of relations is aligned with emerging work in science communication and education emphasizing the role of epistemic affect, how individuals feel about their own knowledge and reasoning (Ogunbode et al., 2022; Sinatra et al., 2014). This mechanism is especially important in scientific topics of social relevance such as CC, where personal relevance, trust in evidence, and critical engagement must coexist for informed civic participation (Boyd et al., 2021; Busch, 2016). These findings contribute to broader discussions in science communication and learning by emphasizing the need for public reasoning strategies that integrate instructional scaffolds, particularly in contexts of uncertainty and value-based conflict (Baram-Tsabari and Lewenstein, 2017; Kranz et al., 2025). Future research could explore communication and instructional strategies that promote epistemic humility and critical reappraisal, especially among learners who consistently engage desired outcome-directed reasoning (Erduran and Dagher, 2014).

An unexpected finding was that confidence in accomplishing sustainability-related tasks did not differ across reasoning groups. This pattern may reflect overconfidence; high certainty not matched by knowledge (Cramer et al., 2009; Said et al., 2023; Stanovich et al., 2013), a phenomenon with important implications for decision-making under uncertainty in climate contexts (Stanton and Roelich, 2021). While this pattern may reflect a form of overconfidence, it also underscores a broader challenge in sustainability communication and education; helping individuals to shift toward a more scientific stance about CC (i.e., judging scientific explanations about the causes of current CC to have greater plausibility than alternative, non-scientific claims). Without structured opportunities to explicitly evaluate the plausibility of explanations, confidence may become decoupled from knowledge and reasoning quality, potentially leading individuals to rely on perceived certainty and a desired outcome rather than critical evaluation when reasoning about complex scientific issues that are socially relevant (Sinatra and Lombardi, 2020; Trémolière and Djeriouat, 2021). This finding also has implications for how confidence is shaped by science outreach and informal learning experiences, suggesting the need for further investigation into how science communicators and out of school experiences might calibrate certainty with reasoning directed toward accuracy (Zummo et al., 2021). This calls for future research to explore how confidence interacts with potentially related constructs, such as metacognitive monitoring and epistemic trust, and whether it serves as a reliable predictor of accuracy-directed reasoning in the presence of misinformation and desired outcome-directed biases.

The study is also among the first to empirically investigate the link between confidence and plausibility judgments regarding CC. By doing so, it opens new avenues for conceptualizing how perceived certainty may (or may not) contribute to the evaluation of scientific explanations about climate. Given widespread overconfidence in public science discourse, future work should examine how confidence calibration –the alignment of confidence with actual knowledge or reasoning quality– can be supported through communication and educational interventions.

The study also contributes to the literature by examining the multifaceted relations between self-efficacy, interest, plausibility judgments, knowledge, and reasoning on sustainability tasks. The study findings shed light on the factors underlying taking action against CC and offer several actionable insights for climate communication and education. First, they reinforce calls to go beyond knowledge transmission and prioritize student engagement with complex issues and scientific reasoning skills (Newton and Annetta, 2025; Zeidler and Sadler, 2023). Communicators and educators can promote situational interest through personally relevant and socially meaningful climate topics, which in turn may stimulate deeper reasoning and sustained involvement (Renninger and Hidi, 2022; Renninger et al., 2023; Wang et al., 2025). Our findings suggest that people’s reasoning and conceptual evaluations may be shaped by how CC is framed, discussed, and problematized in everyday contexts, and highlight the importance of creating collaborative and dialogic spaces in which individuals can examine evidence, express uncertainty, and reflect on values toward greater consensus and understanding; conditions that are likely to support more enduring forms of scientific reasoning and engagement (Governor et al., 2025).

Second, self-efficacy must be explicitly supported in climate communication and education through scaffolded strategies, mastery-based experiences, and peer modeling (Jankowski et al., 2024; Nelson et al., 2022). When people believe they can make a meaningful impact, they are more likely to engage critically and persistently with difficult topics (Hamann and Reese, 2020). Finally, communicators and educators should be equipped not just as content experts, but as reasoning facilitators, able to guide individuals in evaluating claims, navigating uncertainty, and resisting misinformation (Dawson and Carson, 2020; Lombardi et al., 2024). Developing these skills is essential for better developing a more scientifically literate society capable of informed action in response to climate challenges. An important goal is to support accuracy-directed reasoning that guides people to evaluate competing explanations, navigate uncertainty, and connect knowledge to environmentally-responsible agency and action. It is not only about acquiring knowledge; but also reasoning with it toward the goal of accuracy. Scaffolded strategies may help learners and the public to more systematically encourage epistemic reflections, evaluations, and judgments, and reduce reliance on surface-level confidence or identity-based reasoning (Bailey et al., 2022; McGrew, 2024).

5.2 Limitations and future directions

This study has several limitations that merit consideration. First, the cross-sectional design restricts causal inferences about the relationships among variables. Accordingly, the structural models should be interpreted as representations of covariation among constructs rather than evidence of directional or developmental processes. Future longitudinal or experimental work is needed to examine whether the observed patterns of association are stable over time or sensitive to instructional or communicative interventions. For example, longitudinal designs may help clarify whether improvements in epistemic evaluations and judgments are followed by sustained gains in behavior or predict them, and more accuracy-directed decision-making over time. Second, while the study captured motivational and cognitive factors, it did not include behavioral outcome measures. While this study offers valuable insight into the psychological and epistemic conditions that shape readiness for climate action, future research should investigate how cognitive and motivational profiles translate into real-world environmental actions. Finally, although the sample was diverse in age and ethnicity, replicating the study with other demographic and educational groups would strengthen the validity and applicability of these findings.

Furthermore, the reliability of Part B of the RASSI scale (ω = 0.68) fell slightly below conventional thresholds. While retained for exploratory purposes, we acknowledge that lower reliability poses a greater degree of uncertainty in multivariate analyses such as SEM, and future work should focus on refining this measure to strengthen its applicability.

In view of these limitations, the findings of this study remain highly pertinent for the contemporary climate education and communication landscape. As the climate crisis accelerates and public discourse becomes increasingly fragmented by misinformation, educational efforts must focus not only on disseminating scientific knowledge but also on cultivating the motivational and epistemic skills required for scientifically informed engagement. This study underscores the pivotal role of plausibility judgments as a mediating construct in the observed relations between self-efficacy, interest, confidence, reasoning, and knowledge construction. By emphasizing these factors, climate education and communication can more effectively facilitate a transition from passive awareness to active reasoning and sustainable action among learners and the general public. When considering and acting on scientific issues of social relevance, such as CC, our findings highlight the value of studying plausibility as one form of epistemic judgment, alongside credibility judgments about source trustworthiness, expertise, and reliability (Rescher, 2006; Reynolds and McGrew, 2025). The integration of motivational and cognitive support into communication and instructional design serves as a powerful catalyst for cultivating scientifically grounded, accuracy-directed reasoning and more effective decision-making in response to the intricate challenges posed by CC.

References

• 1

  AndreT.WindschitlM. (2003). “Interest, epistemological belief, and intentional conceptual change” in Intentional conceptual change (New York, NY: Routledge), 179–203.

  • Google Scholar

• 2

  Azila-GbettorE. M.MensahC.AbiemoM. K.BokorM. (2021). Predicting student engagement from self-efficacy and autonomous motivation: a cross-sectional study. Cogent Educ.8:1942638. doi: 10.1080/2331186X.2021.1942638

  • CrossRef
  • Google Scholar

• 3

  BaderJ. D.AhearnK. A.AllenB. A.AnandD. M.CoppensA. D.AikensM. L. (2023). The decision is in the details: justifying decisions about socioscientific issues. J. Res. Sci. Teach.60, 2147–2179. doi: 10.1002/tea.21854

  • CrossRef
  • Google Scholar

• 4

  BaileyJ. M.HanedarM.SinatraG. M.McAuliffeC.SchouteE. C.BrighterK. (2025). Developing and validating a multi-tiered instrument for assessing knowledge and reasoning about socioscientific issues. Discip. Interdiscip. Sci. Educ. Res.

  • Google Scholar

• 5

  BaileyJ. M.JamaniS.KlavonT. G.JaffeJ.MohanS. (2022). Climate crisis learning through scaffolded instructional tools. Educ. Dev. Psychol.39, 85–99. doi: 10.1080/20590776.2021.1997065

  • CrossRef
  • Google Scholar

• 6

  BaileyJ. M.LombardiD.CordovaJ. R.SinatraG. M. (2017). Meeting students halfway: increasing self-efficacy and promoting knowledge change in astronomy. Phys. Rev. Phys. Educ. Res.13:020140. doi: 10.1103/PhysRevPhysEducRes.13.020140

  • CrossRef
  • Google Scholar

• 7

  BanduraA. (1997). Self-efficacy: the exercise of control, vol. 11. New York, NY: W. H. Freeman and Company.

  • Google Scholar

• 8

  BanduraA.CherryL. (2020). Enlisting the power of youth for climate change. Am. Psychol.75, 945–951. doi: 10.1037/amp0000512,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  Baram-TsabariA.LewensteinB. V. (2017). Science communication training: what are we trying to teach?Int. J. Sci. Educ., Part B7, 285–300. doi: 10.1080/21548455.2017.1303756

  • CrossRef
  • Google Scholar

• 10

  BesleyJ. C.DownsM. R. (2025). Environmental scientists’ support for public engagement strategy development is predicted by a range of factors, but mostly perceived benefits. Int. J. Sci. Educ., Part B15, 1–16. doi: 10.1080/21548455.2024.2342039

  • CrossRef
  • Google Scholar

• 11

  BoydK. J.OkochiC.LittrellM. K.BatchelorR. L.GoldA. U.LeckeyE. (2021). Career implications for college student mentors through their contribution to a climate film outreach program for adolescents. Int. J. Sci. Educ., Part B11, 308–323. doi: 10.1080/21548455.2021.1996656

  • CrossRef
  • Google Scholar

• 12

  BuschK. C. (2016). Polar bears or people? Exploring ways in which teachers frame climate change in the classroom. Int. J. Sci. Educ., Part B6, 137–165. doi: 10.1080/21548455.2015.1027320

  • CrossRef
  • Google Scholar

• 13

  CapassoM.GuidettiM.BianchiM.CavazzaN.CasoD. (2025). Enhancing intentions to reduce meat consumption: an experiment comparing the role of self-and social pro-environmental identities. J. Environ. Psychol.101:102494. doi: 10.1016/j.jenvp.2024.102494

  • CrossRef
  • Google Scholar

• 14

  ChenP.BaoC.GaoQ. (2021). Proactive personality and academic engagement: the mediating effects of teacher-student relationships and academic self-efficacy. Front. Psychol.12:652994. doi: 10.3389/fpsyg.2021.652994

  • CrossRef
  • Google Scholar

• 15

  ChinnC. A.BucklandL. A.SamarapungavanA. L. A. (2011). Expanding the dimensions of epistemic cognition: arguments from philosophy and psychology. Educ. Psychol.46, 141–167. doi: 10.1080/00461520.2011.587722

  • CrossRef
  • Google Scholar

• 16

  ChoungH.NewmanT. P.StenhouseN. (2020). The role of epistemic beliefs in predicting citizen interest and engagement with science and technology. Int. J. Sci. Educ., Part B10, 248–265. doi: 10.1080/21548455.2020.1774094

  • CrossRef
  • Google Scholar

• 17

  CramerR. J.NealT.BrodskyS. L. (2009). Self-efficacy and confidence: theoretical distinctions and implications for trial consultation. Consult. Psychol. J. Pract. Res.61, 319–334. doi: 10.1037/a0017310

  • CrossRef
  • Google Scholar

• 18

  DawsonV.CarsonK. (2020). Introducing argumentation about climate change socioscientific issues in a disadvantaged school. Res. Sci. Educ.50, 863–883. doi: 10.1007/s11165-018-9715-x

  • CrossRef
  • Google Scholar

• 19

  DeVellisR. F.ThorpeC. T. (2021). Scale development: theory and applications. Thousand Oaks, CA: Sage Publications.

  • Google Scholar

• 20

  DittoP. H.LiuB. S.ClarkC. J.WojcikS. P.ChenE. E.GradyR. H.et al. (2019). At least bias is bipartisan: a meta-analytic comparison of partisan bias in liberals and conservatives. Perspect. Psychol. Sci.14, 273–291. doi: 10.1177/1745691617746796,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21

  DoleJ. A.SinatraG. M. (1998). Reconceptalizing change in the cognitive construction of knowledge. Educ. Psychol.33, 109–128. doi: 10.1080/00461520.1998.9653294

  • CrossRef
  • Google Scholar

• 22

  DuitR.TreagustD. F. (2003). Conceptual change: a powerful framework for improving science teaching and learning. Int. J. Sci. Educ.25, 671–688. doi: 10.1080/09500690305016

  • CrossRef
  • Google Scholar

• 23

  ErduranS.DagherZ. R. (2014). “Reconceptualizing nature of science for science education” in Reconceptualizing the nature of science for science education: Scientific knowledge, practices and other family categories (Dordrecht, Netherlands: Springer), 1–18.

  • Google Scholar

• 24

  FaulF.ErdfelderE.BuchnerA.LangA. G. (2009). Statistical power analyses using G* power 3.1: tests for correlation and regression analyses. Behav. Res. Methods41, 1149–1160. doi: 10.3758/BRM.41.4.1149,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  FawzyS.OsmanA. I.DoranJ.RooneyD. W. (2020). Strategies for mitigation of climate change: a review. Environ. Chem. Lett.18, 2069–2094. doi: 10.1007/s10311-020-01059-w

  • CrossRef
  • Google Scholar

• 26

  FornellC.LarckerD. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. J. Mark. Res.18, 39–50. doi: 10.1177/002224378101800104

  • CrossRef
  • Google Scholar

• 27

  GeigerN.PasekM. H.GruszczynskiM.RatcliffN. J.WeaverK. S. (2020). Political ingroup conformity and pro-environmental behavior: evaluating the evidence from a survey and mousetracking experiments. J. Environ. Psychol.72:101524. doi: 10.1016/j.jenvp.2020.101524

  • CrossRef
  • Google Scholar

• 28

  GoldsteinC. M.MurrayE. J.BeardJ.SchnoesA. M.WangM. L. (2020). Science communication in the age of misinformation. Ann. Behav. Med.54, 985–990. doi: 10.1093/abm/kaaa088,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29

  GovernorD.McAuliffeC.Ramirez VillarinL.KlavonT. G.van MeertenJ. E.RachelD.et al. (2025). Facilitating student argumentation around socioscientific issues through productive discourse and negotiation toward consensus. Theory Pract.64, 168–184. doi: 10.1080/00405841.2025.2453367

  • CrossRef
  • Google Scholar

• 30

  HamannK. R.ReeseG. (2020). My influence on the world (of others): goal efficacy beliefs and efficacy affect predict private, public, and activist pro-environmental behavior. J. Soc. Issues76, 35–53.

  • Google Scholar

• 31

  HanedarM.OzyaziciG.CeyhanG. D. (2024). Assessment and validation: an updated climate change plausibility perception measure. Environ. Educ. Res.30, 2291–2314. doi: 10.1080/13504622.2024.2341172

  • CrossRef
  • Google Scholar

• 32

  HerrickI. R.SinatraG. M.LombardiD. (2023). Is that plausible?Sci. Teach.90, 55–59. doi: 10.1080/00368555.2023.12294208

  • CrossRef
  • Google Scholar

• 33

  HidiS.RenningerK. A. (2006). The four-phase model of interest development. Educ. Psychol.41, 111–127. doi: 10.1207/s15326985ep4102_4

  • CrossRef
  • Google Scholar

• 34

  HonickeT.BroadbentJ. (2016). The influence of academic self-efficacy on academic performance: a systematic review. Educ. Res. Rev.17, 63–84. doi: 10.1016/j.edurev.2015.11.002

  • CrossRef
  • Google Scholar

• 35

  IPCC (2022). “Climate change 2022: impacts, adaptation, and vulnerability” in Contribution of working group II to the sixth assessment report of the intergovernmental panel on climate change. eds. PörtnerH.-O.RobertsD. C.TignorM.PoloczanskaE. S.MintenbeckK.AlegríaA.et al. (Cambridge: Cambridge University Press).

  • Google Scholar

• 36

  IPCC (2023). “Climate change 2023: synthesis report” in Contribution of working groups I, II and III to the sixth assessment report of the intergovernmental panel on climate change. eds. TeamC. W.LeeH.RomeroJ. (Geneva, Switzerland: IPCC), 35–115.

  • Google Scholar

• 37

  JankowskiJ. M.MlynskiC.JobV. (2024). Just a drop in the ocean? How lay beliefs about the world influence efficacy, perceptions, and intentions regarding pro-environmental behavior. J. Environ. Psychol.100:102445. doi: 10.1016/j.jenvp.2024.102445

  • CrossRef
  • Google Scholar

• 38

  JASP Team, 2024. JASP (version 0.95.0) [computer software]. Available online at: https://jasp-stats.org (Accessed June 3, 2025).

  • Google Scholar

• 39

  JugertP.GreenawayK. H.BarthM.BüchnerR.EisentrautS.FritscheI. (2016). Collective efficacy increases pro-environmental intentions through increasing self-efficacy. J. Environ. Psychol.48, 12–23. doi: 10.1016/j.jenvp.2016.08.003

  • CrossRef
  • Google Scholar

• 40

  KahanD. M. (2013). Ideology, motivated reasoning, and cognitive reflection. Judgm. Decis. Mak.8, 407–424. doi: 10.1017/S1930297500005271

  • CrossRef
  • Google Scholar

• 41

  KahanD. M.PetersE.DawsonE. C.SlovicP. (2017). Motivated numeracy and enlightened self-government. Behav. Public Policy1, 54–86. doi: 10.1017/bpp.2016.2

  • CrossRef
  • Google Scholar

• 42

  KockN. (2017). WarpPLS user manual: version 6.0., vol. 141. Laredo: ScriptWarp Systems, 47–60.

  • Google Scholar

• 43

  KockN.LynnG. S. (2012). Lateral collinearity and misleading results in variance-based SEM: an illustration and recommendations. J. Assoc. Inf. Syst.13, 546–580. doi: 10.17705/1jais.00302

  • CrossRef
  • Google Scholar

• 44

  KranzJ.BreitenmoserP.LahertoA.KrugA.SchwichowM.TasquierG. (2025). Science education for collective agency in the climate crisis: a social identity approach. Res. Sci. Educ.55, 1149–1168. doi: 10.1007/s11165-025-10282-w

  • CrossRef
  • Google Scholar

• 45

  KundaZ. (1990). The case for motivated reasoning. Psychol. Bull.108, 480–498. doi: 10.1037/0033-2909.108.3.480,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  LaurenN.FieldingK. S.SmithL.LouisW. R. (2016). You did, so you can and you will: self-efficacy as a mediator of spillover from easy to more difficult pro-environmental behaviour. J. Environ. Psychol.48, 191–199. doi: 10.1016/j.jenvp.2016.10.004

  • CrossRef
  • Google Scholar

• 47

  LevyB. L.OliveiraA. W.HarrisC. B. (2021). The potential of “civic science education”: theory, research, practice, and uncertainties. Sci. Educ.105, 1053–1075. doi: 10.1002/sce.21678

  • CrossRef
  • Google Scholar

• 48

  LinnenbrinkE. A.PintrichP. R. (2002). Motivation as an enabler for academic success. Sch. Psychol. Rev.31, 313–327. doi: 10.1080/02796015.2002.12086158

  • CrossRef
  • Google Scholar

• 49

  LombardiD.SinatraG. M. (2012). College students’ perceptions about the plausibility of human-induced climate change. Res. Sci. Educ.42, 201–217. doi: 10.1007/s11165-010-9196-z

  • CrossRef
  • Google Scholar

• 50

  LombardiD.NussbaumE. M.SinatraG. M. (2016). Plausibility judgments in conceptual change and epistemic cognition. Educ. Psychol.51, 35–56. doi: 10.1080/00461520.2015.1113134

  • CrossRef
  • Google Scholar

• 51

  LombardiD.SinatraG. M.NussbaumE. M. (2013). Plausibility reappraisals and shifts in middle school students' climate change conceptions. Learn. Instr.27, 50–62. doi: 10.1016/j.learninstruc.2013.03.001

  • CrossRef
  • Google Scholar

• 52

  LombardiD.SinatraG. M.BaileyJ. M.ButlerL. P. (2024). Seeking a comprehensive theory about the development of scientific thinking. Educ. Psychol. Rev.36:72. doi: 10.1007/s10648-024-09911-z

  • CrossRef
  • Google Scholar

• 53

  MartinieM. A.ShanklandR. (2024). Achievement goals, self-efficacy, and psychological flexibility as antecedent of study engagement. Soc. Psychol. Educ.27, 2395–2416. doi: 10.1007/s11218-024-09921-3

  • CrossRef
  • Google Scholar

• 54

  MasonL.GavaM.BoldrinA. (2008). On warm conceptual change: the interplay of text, epistemological beliefs, and topic interest. J. Educ. Psychol.100:291. doi: 10.1037/0022-0663.100.2.291

  • CrossRef
  • Google Scholar

• 55

  McGrewS. (2024). Teaching lateral reading: interventions to help people read like fact checkers. Curr. Opin. Psychol.55:101737. doi: 10.1016/j.copsyc.2023.101737

  • CrossRef
  • Google Scholar

• 56

  MengQ.ZhangQ. (2023). The influence of academic self-efficacy on university students’ academic performance: the mediating effect of academic engagement. Sustainability15:5767. doi: 10.3390/su15075767

  • CrossRef
  • Google Scholar

• 57

  NelsonS. M.IraG.MerenlenderA. M. (2022). Adult climate change education advances learning, self-efficacy, and agency for community-scale stewardship. Sustainability14:1804. doi: 10.3390/su14031804

  • CrossRef
  • Google Scholar

• 58

  NewtonM. H.AnnettaL. A. (2025). The influence of extended reality on climate change education. Sci. & Educ.34, 825–845. doi: 10.1007/s11191-024-00518-y

  • CrossRef
  • Google Scholar

• 59

  NunnallyJ. C. (1978). Psychometric theory. 2nd Edn. New York, NY: McGraw-Hill.

  • Google Scholar

• 60

  OgunbodeC. A.DoranR.HanssD.OjalaM.Salmela-AroK.van den BroekK. L.et al. (2022). Climate anxiety, pro-environmental action and wellbeing: antecedents and outcomes of negative emotional responses to climate change in 28 countries. J. Environ. Psychol.84:1. doi: 10.1016/j.jenvp.2022.101887

  • CrossRef
  • Google Scholar

• 61

  PajaresF. (2002). Gender and perceived self-efficacy in self-regulated learning. Theory Pract.41, 116–125. doi: 10.1207/s15430421tip4102_8

  • CrossRef
  • Google Scholar

• 62

  PennycookG.RandD. G. (2019). Lazy, not biased: susceptibility to partisan fake news is better explained by lack of reasoning than by motivated reasoning. Cognition188, 39–50. doi: 10.1016/j.cognition.2018.06.011,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 63

  PetersonR. A. (1994). A meta-analysis of Cronbach's coefficient alpha. J. Consum. Res., 381–391. https://www.jstor.org/stable/2489828

  • Google Scholar

• 64

  PosnerG. J.StrikeK. A.HewsonP. W.GertzogW. A. (1982). Accommodation of a scientific conception: toward a theory of conceptual change. Sci. Educ.66, 211–227. doi: 10.1002/sce.3730660207

  • CrossRef
  • Google Scholar

• 65

  PreacherK. J.HayesA. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav. Res. Methods40, 879–891. doi: 10.3758/BRM.40.3.879,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 66

  ReeseG.JungeE. A. (2017). Keep on rockin’in a (plastic-) free world: collective efficacy and pro-environmental intentions as a function of task difficulty. Sustainability9:200. doi: 10.3390/su9020200

  • CrossRef
  • Google Scholar

• 67

  RenningerK. A.HidiS. E. (2022). Interest development, self-related information processing, and practice. Theory Pract.61, 23–34. doi: 10.1080/00405841.2021.1932159

  • CrossRef
  • Google Scholar

• 68

  RenningerK. A.HidiS. E.DeA. (2023). “Exploring interest theory and its reciprocal relation to achievement goals, self-efficacy, and self-regulation” in Motivation and emotion in learning and teaching across educational contexts (New York, NY: Routledge), 19–34.

  • Google Scholar

• 69

  RescherN. (2006). Presumption and the practices of tentative cognition. Cambridge, UK: Cambridge University Press.

  • Google Scholar

• 70

  ReynoldsE. C.McGrewS. (2025). “From personal experience, it’s true!”: students’ digital evaluations on relevant issues. J. Learn. Sci.34, 1–44. doi: 10.1080/10508406.2025.2495214

  • CrossRef
  • Google Scholar

• 71

  RippleW. J.WolfC.GreggJ. W.RockströmJ.MannM. E.OreskesN.et al. (2024). The 2024 state of the climate report: perilous times on planet earth. Bioscience74, 812–824. doi: 10.1093/biosci/biae087

  • CrossRef
  • Google Scholar

• 72

  SaidN.FrauhammerL. T.HuffM. (2023). Consensus messaging in climate change communication: metacognition as moderator variable in the gateway belief model. J. Environ. Psychol.91:102128. doi: 10.1016/j.jenvp.2023.102128

  • CrossRef
  • Google Scholar

• 73

  SarstedtM.RingleC. M.HairJ. F. (2021). “Partial least squares structural equation modeling” in Handbook of market research (Cham: Springer International Publishing), 587–632.

  • Google Scholar

• 74

  SinatraG. M. (2005). The" warming trend" in conceptual change research: the legacy of Paul R. Pintrich. Educational psychologist40, 107–115. doi: 10.1207/s15326985ep4002_5

  • CrossRef
  • Google Scholar

• 75

  SinatraG. M.LombardiD. (2020). Evaluating sources of scientific evidence and claims in the post-truth era may require reappraising plausibility judgments. Educ. Psychol.55, 120–131. doi: 10.1080/00461520.2020.1730181

  • CrossRef
  • Google Scholar

• 76

  SinatraG. M.KienhuesD.HoferB. K. (2014). Addressing challenges to public understanding of science: epistemic cognition, motivated reasoning, and conceptual change. Educ. Psychol.49, 123–138. doi: 10.1080/00461520.2014.916216

  • CrossRef
  • Google Scholar

• 77

  SmithM.FinklerW.AitkenR. (2023). Connecting people with science: a proof-of-concept study to evaluate action-based storytelling for science communication. Sustainability15:11655. doi: 10.3390/su151511655

  • CrossRef
  • Google Scholar

• 78

  StanovichK. E.WestR. F.ToplakM. E. (2013). Myside bias, rational thinking, and intelligence. Curr. Dir. Psychol. Sci.22, 259–264. doi: 10.1177/0963721413480174

  • CrossRef
  • Google Scholar

• 79

  StantonM. C. B.RoelichK. (2021). Decision making under deep uncertainties: a review of the applicability of methods in practice. Technol. Forecast. Soc. Chang.171:120939. doi: 10.1016/j.techfore.2021.120939

  • CrossRef
  • Google Scholar

• 80

  SyropoulosS.MarkowitzE. M. (2022). Perceived responsibility to address climate change consistently relates to increased pro-environmental attitudes, behaviors and policy support: evidence across 23 countries. J. Environ. Psychol.83:101868. doi: 10.1016/j.jenvp.2022.101868

  • CrossRef
  • Google Scholar

• 81

  TaberneroC.HernándezB. (2011). Self-efficacy and intrinsic motivation guiding environmental behavior. Environ. Behav.43, 658–675. doi: 10.1177/0013916510379759

  • CrossRef
  • Google Scholar

• 82

  Teshera-LevyeJ.AlamI.CorwinL.DauerJ. M. (2025). Developing a conceptual model for students’ scientific civic engagement self-concept. Int. J. Sci. Educ., 1–22. doi: 10.1080/09500693.2024.2449220

  • CrossRef
  • Google Scholar

• 83

  TrémolièreB.DjeriouatH. (2021). Exploring the roles of analytic cognitive style, climate science literacy, illusion of knowledge, and political orientation in climate change skepticism. J. Environ. Psychol.74:101561. doi: 10.1016/j.jenvp.2021.101561

  • CrossRef
  • Google Scholar

• 84

  VulturiusG.Gerger SwartlingÅ. (2015). Overcoming social barriers to learning and engagement with climate change adaptation: experiences with Swedish forestry stakeholders. Scand. J. For. Res.30, 217–225. doi: 10.1080/02827581.2014.1002218

  • CrossRef
  • Google Scholar

• 85

  WallisH.LoyL. S. (2021). What drives pro-environmental activism of young people? A survey study on the Fridays for future movement. J. Environ. Psychol.74:101581. doi: 10.1016/j.jenvp.2021.101581

  • CrossRef
  • Google Scholar

• 86

  WangC.GengL.CasallasJ. D. R. (2022). Mindfulness to climate change inaction: the role of awe, “dragons of inaction” psychological barriers and nature connectedness. J. Environ. Psychol.84:101912. doi: 10.1016/j.jenvp.2022.101912

  • CrossRef
  • Google Scholar

• 87

  WangX.ChenC.QiJ.ChenK.HuangP.LuoC.et al. (2025). Positive emotional experiences from nature contact: a thematic analysis on types and regulation pathways. J. Environ. Psychol.103:102561. doi: 10.1016/j.jenvp.2025.102561

  • CrossRef
  • Google Scholar

• 88

  WangY.ZhangW. (2024). The relationship between college students’ learning engagement and academic self-efficacy: a moderated mediation model. Front. Psychol.15:1425172. doi: 10.3389/fpsyg.2024.1425172

  • CrossRef
  • Google Scholar

• 89

  WassersteinR. L.SchirmA. L.LazarN. A. (2019). Moving to a world beyond “p< 0.05”. Am. Stat.73, 1–19. doi: 10.1080/00031305.2019.1583913

  • CrossRef
  • Google Scholar

• 90

  WestS. G.FinchJ. F.CurranP. J. (1995). “Structural equation models with nonnormal variables” in Structural equation modeling: Concepts, issues, and applications. ed. HoyleR. H. (Thousand Oaks, CA: Sage), 56–75.

  • Google Scholar

• 91

  WetzelsM.Odekerken-SchröderG.Van OppenC. (2009). Using PLS path modeling for assessing hierarchical construct models: guidelines and empirical illustration. MIS Q.33, 177–195. doi: 10.2307/20650284

  • CrossRef
  • Google Scholar

• 92

  WilkinsonC.MilaniE.RidgwayA.WeitkampE. (2023). Motivations and deterrents in contemporary science communication: a questionnaire survey of actors in seven European countries. Int. J. Sci. Educ., Part B13, 131–148. doi: 10.1080/21548455.2022.2139165

  • CrossRef
  • Google Scholar

• 93

  ZeidlerD. L.NicholsB. H. (2009). Socioscientific issues: theory and practice. J. Elem. Sci. Educ.21, 49–58. doi: 10.1007/BF03173684

  • CrossRef
  • Google Scholar

• 94

  ZeidlerD. L.SadlerT. D. (2023). “Exploring and expanding the frontiers of socioscientific issues” in Handbook of research on science education (New York, NY: Routledge), 899–929.

  • Google Scholar

• 95

  ZeidlerD. L.HermanB. C.SadlerT. D. (2019). New directions in socioscientific issues research. Discip. Interdiscip. Sci. Educ. Res.1, 1–9. doi: 10.1186/s43031-019-0008-7

  • CrossRef
  • Google Scholar

• 96

  ZhangJ.CaoA. (2025). The psychological mechanisms of education for sustainable development: environmental attitudes, self-efficacy, and social norms as mediators of pro-environmental behavior among university students. Sustainability17:933. doi: 10.3390/su17030933

  • CrossRef
  • Google Scholar

• 97

  ZhaoX.LynchJ. G.Jr.ChenQ. (2010). Reconsidering baron and Kenny: myths and truths about mediation analysis. J. Consum. Res.37, 197–206. doi: 10.1086/651257

  • CrossRef
  • Google Scholar

• 98

  ZummoL.GargroetziE.GarciaA. (2021). How youth use scientific argumentation in civic participation on climate change: polar bears, the great barrier reef, and ‘your job as president’. Int. J. Sci. Educ., Part B11, 362–377. doi: 10.1080/21548455.2021.2010830

  • CrossRef
  • Google Scholar

• 99

  ZysbergL.SchwabskyN. (2021). School climate, academic self-efficacy and student achievement. Educ. Psychol.41, 467–482. doi: 10.1080/01443410.2020.1813690

  • CrossRef
  • Google Scholar