On the Strength of Trust: The Moderating Role of Certainty in Judgments of Authorities

Abstract

Communication studies often measure trust in or credibility of message sources as if their effects on critical consequences such as behavior are uniform across individuals and contexts. Drawing on the literature of attitude strength, this paper argues that considering the certainty of trust judgments enables researchers to better understand when trust is likely to induce desired behavior such as cooperating with a judged authority. Using two studies in different contexts, one from a local environmental dispute and another from the national-level COVID-19 pandemic in the U.S., the current paper shows that communication practices (e.g., personal interaction, news media use, interpersonal discussion) reinforce people’s certainty in their judgment of authorities. This certainty, in turn, interacts with the effect of trust such that trust judgments greater in certainty are more likely to lead to behavioral intentions.

Keywords

trust topics survey method experiment method

Communication research has long acknowledged how people’s judgments of authorities influence persuasion outcomes. For example, the literature on source credibility, which refers to “judgments made by a perceiver concerning the believability of a communicator” (O’Keefe, 2015, p. 291), finds that sources with higher credibility induce more attitude change and behavioral compliance (Pornpitakpan, 2004). Similarly, communication subdisciplines focusing on the relationship between authorities and the public including risk communication have sought to understand how trust predicts critical outcomes such as risk perception, technology acceptance, and support for policy decisions (Siegrist, 2021). In these relationships, members of the public overwhelmed by the complexity of new environmental, technological, and health risks often find it easier to judge the authorities managing those risks instead (Earle & Cvetkovich, 1995). Although this replacement partially relieves individuals from cognitive burden, it exposes them to the malfeasance or negligence of the authorities, a new kind of risk. Accordingly, research has conceptualized trust as a psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions or behavior of another (Rousseau et al., 1998, p. 395) and many studies have operationalized trust as the public’s evaluative judgment of an authority akin to credibility assessments (Allum, 2007; Earle et al., 2007; Frewer et al., 1996; Hunt et al., 1999; McComas & Trumbo, 2001; Peters et al., 1997; Siegrist et al., 2012).

Evaluative judgments of a source’s credibility can be considered as attitudes in the sense that an attitude is a “psychological tendency that is expressed by evaluating a particular entity with some degree of favor or disfavor” (Eagly & Chaiken, 1993, p. 1). In a similar vein, although some perspectives argue that trust is a choice reflecting one’s decision to accept vulnerability by relying on another (trust-as-choice), majority of empirical studies in communication treat trust as the trustor’s internalized willingness to accept this vulnerability arising from the perceived trustworthiness of the party under judgment (trust-as-attitude) (Li, 2007). To the extent that the terms credibility and trust reflect an evaluative judgment of public authorities as an information source, they are used interchangeably in this paper although trust is featured more frequently due to the term’s prevalent use in risk-related contexts from which this paper derived and tested its hypotheses.

Considering trust judgments as a special type of attitudes, findings from attitude research can be adapted to generate new insights to improve theorizing of trust. Conventional approaches to measuring trust or credibility have rarely considered how or why the effect of these judgments vary across populations or contexts. For example, people may trust an authority because they watched her for many years working with rigor, integrity, and care. Others may trust the same authority because they simply assume that authorities in general come into their position after some unknown vetting process that verifies their qualifications. It is not difficult to imagine which of these two forms of trust would more likely lead to intentions to follow authorities’ advice, but most measures of trust have disregarded or failed to capture these nuances indicative of people’s strength of trust.

This paper first seeks to reveal the influence different communication practices have in shaping the certainty with which individuals hold their trust judgments. Then, it shows that research can better predict when trust leads to consequential outcomes (i.e., behavioral intentions) by considering the role of certainty of trust judgments as a moderator. To this end, two empirical studies conducted in distinct contexts, namely a survey based on a local environmental dispute and an online experiment conducted during the COVID-19 pandemic, test hypotheses derived from attitude certainty and communication research.

Literature Review

Conflicting Views About Attitudes

At least two different perspectives exist among attitude researchers regarding the ontological status of attitudes (Bohner & Dickel, 2010). On the one hand, some scholars hold that attitudes are improvised constructions, in other words evaluative judgments formed using feelings or thoughts most salient and accessible at the time when such judgments are needed (Conrey & Smith, 2007; Gawronski & Bodenhausen, 2007; Schwarz, 2007). According to this perspective, attitudes are the result of affective responses depending on the mental associations activated when we encounter an attitude object (Gawronski & Bodenhausen, 2007). For this reason, subtle and trivial influences such as changes in question wordings (Schuldt et al., 2011; Schuman et al., 1981; Schwarz, 1999) or different states of mood across time (Isen et al., 1978; Schwarz et al., 1987) can sway our evaluative judgments.

On the other hand, other researchers argue that evaluative judgments occur because people retrieve stored attitudes from memory (Fazio, 2007; Petty et al., 2007). According to this view, attitudes are associations between attitude objects and attached evaluations represented in memory varying in strength and activated upon encounter of the object (Fazio, 2007). As such, attitudes are simply a special type of knowledge associated with evaluative information summarizing past experiences with an object, available for retrieval like file drawers when needed. This perspective better explains how attitudes with familiar objects remain stable (Fazio, 2007).

Inconsistent Assumptions in Trust Research

These competing perspectives toward attitudes invite researchers to reflect on their assumptions when they study trust in or credibility of authorities. Research on trust, when operationalized as attitudes, inevitably aligns with one of these conflicting views but rarely articulates these assumptions save for a few exceptions (e.g., Metzger, 2007; Metzger et al., 2010). Some researchers appear to assume that participants hold specific evaluative judgments based on some information stored in their memories such as when a study asks for judgments on a source’s specific aspects (e.g., honesty, responsibility, care, expertise, competence). Studies exploring the dimensionality of trust and credibility by soliciting evaluations of an authority on these detailed qualities assume that participants hold matching segmented knowledge and effectively view these judgments as stored memory (e.g., Berlo et al., 1969; Earle et al., 2007; Johnson, 1999; McGinnies & Ward, 1980).

In contrast, research aligned with other theoretical perspectives posit that people cannot form and hold knowledge of such qualities in memory due to its large cognitive demand (Earle, 2010). Instead, individuals use heuristics readily available to judge the credibility of authorities even when they have little diagnostic value. Such heuristics may even include physical attractiveness (Patzer, 1983), dynamism (Berlo et al., 1969; Whitehead, 1968), perceived similarity in identity or values (Earle et al., 2007; Siegrist et al., 2000), and affect at the time of judgment which may be unrelated to the risk managed by the authority (Dunn & Schweitzer, 2005). The trust, confidence, and cooperation model of trust (Earle et al., 2007) and its antecedent theory, the salient value similarity model of social trust (Cvetkovich et al., 2002; Siegrist et al., 2000) argue that any values “currently active for the observer” (Earle et al., 2007, p. 9) comprise the criteria by which an authority will be judged on its moral and relational aspects. In a similar vein, studies drawing on Hovland et al.’s (1953) seminal work on source credibility have also operationalized the construct as a heuristic (McComas & Trumbo, 2001; Trumbo, 1999). Assumptions underlying these approaches are consistent with the view that attitudes are constructions established on the spot when the need arises and are at odds with approaches treating trust as stable dispositions.

When Does Trust Become Impactful? An Attitude Strength Perspective

It appears that both of the perspectives reflect alternative aspects of reality regarding the nature of trust judgments. For instance, whereas the view that trust and credibility are stable dispositions can better explain how the effects of trust can be enduring and impactful, the view that these judgments are instant constructions more effectively account for how people can respond to survey items about authorities whom they barely recognize Hunt et al. (1999), Walls et al. (2004). However, scant research in this area has systematically accounted for the conditions that shape these different forms of trust. The literature of attitude strength could provide structural guidance to help decipher when attitudes are retrieved from memory versus constructed on the spot (Bohner & Dickel, 2010; Fazio, 2007; Nayakankuppam et al., 2018; Petty et al., 2007).

According to the definition by Krosnick and Petty (1995), attitude strength is defined as the extent to which an attitude exhibits four features: (1) it remains unchanged over extended period of time (persistence); (2) it withstands counter-attitudinal attacks (resistance); (3) it influences information processing and judgments by making it easier for certain information to come to mind, often leading to bias (impact on information processing); and (4) it guides attitude-consistent behavior (impact on behavior). Petty and Krosnick (1995) also identified and listed attributes of attitude strength predicting these consequences which include importance (Krosnick, 1988; Zuwerink & Devine, 1996), knowledge (Wood et al., 1995), and elaboration (Petty et al., 1995; Wegener et al., 1995). In addition, attitude certainty, the focus of this study, refers to the extent to which one is confident or sure of one’s attitude (Petty & Krosnick, 1995). Thus, an attitude strength approach would predict that attitudes held with greater certainty, for example, will be more likely to last longer stably and induce behavior consistent with the authority’s message. Despite superficial similarities with attitude extremity (i.e., the extent to which an attitude deviates from neutrality; Petty & Krosnick, 1995), research suggests that certainty and extremity are conceptually and empirically distinct from each other (Rucker et al., 2008; Tormala & Rucker, 2007).

Whereas attitude research, through research on attitude strength and dual-processing theories (Chaiken, 1980; Petty & Cacioppo, 1986), have sought to reconcile the two perspectives of attitudes as transient versus stable entities, research in trust and credibility has rarely made corresponding effort. Instead of categorically assuming that trust judgments are either inherently weak or strong, research can directly address and model this variance to improve understanding about the quality of trust. Drawing on findings from two studies, the current paper seeks to understand the role of communication practices as a source of variance in certainty of trust judgments and the moderating effects such certainty has in the relationship between trust and behavioral outcomes indicative of cooperation with the authority.

Attitude Certainty: Antecedents and Consequences

Certainty can facilitate the development of trust when it is bolstered through direct and repeated interactions (Tormala & Rucker, 2007). Research has found that individuals who had direct experience with an attitude object become more confident in their attitudes than those with only indirect experience. This confidence, in turn, predicted strength-defining consequences such as attitude-expressive behavior (Fazio & Zanna, 1978a). In addition, repeated encounters with the target in an ongoing relationship can also increase certainty because repeated expression of an attitude reinforces one’s confidence in the attitude (Holland et al., 2003; Petrocelli et al., 2007).

Antecedents of Trust Judgment Certainty

In light of previous research, to understand the sources of certainty in trust judgments, research must consider how people gain information about authorities, when people are prompted to form judgments about them, and when they become motivated to think deeply about them. Recent theorizing concerning civic engagement recognizes not only that both mass and interpersonal communication processes influence political participation (Cho et al., 2009) but also that these processes can simultaneously strengthen trust and distrust through polarizing discourse (Shah et al., 2017). Accordingly, using variables that capture direct and mediated experiences in which people talk with, observe, or discuss the conduct of authorities can help predict certainty.

First, direct engagement with authorities afford opportunities to learn and think deeply about them as individuals rather than faceless representatives of organizations. Adapting the definition used by Nabatchi and Amsler (2014), the category “direct engagement” here encompasses experiences in which individuals personally observe or interact with risk-managing authority figures without the media or interest groups as intermediaries, either in face-to-face events such as public hearings or through some personalized communication method (e.g., personally addressed letters, phone calls, or emails). Communication studies have traditionally regarded direct, face-to-face interactions with a special status in that they enable exchange of rich information between involved parties. Compared to conversation mediated through telecommunication technologies or written text, face-to-face interaction affords richer social information that helps individuate a communication partner (Tanis & Postmes, 2003) which leads to greater perceptions of likability and intelligence (Fullwood, 2007). This richness of information enhances clarity and reduces ambiguities in communication, improving collaborative experiences and decision outcomes among participants in deliberative settings relative to mediated modes of communication (Baek et al., 2012; Olson & Olson, 2000). When judging authorities through face-to-face interaction, individuals may detect and use subtle cues signaling unarticulated intentions such as condescending smirks or displays of reluctance to disclose information. In a similar vein, attitude strength research suggests that direct experience with an attitude object is likely to increase the amount of information available about the object and the accessibility of the attitude, enabling individuals to hold strong attitudes (Fazio et al., 1982; Fazio & Zanna, 1981).

H1: Direct engagement with authorities will be positively related to the certainty of trust judgments about authorities.

Second, use of news media can predict the certainty of one’s judgments about authorities. Although exposure to mass-mediated information about an authority may not be as impactful as direct engagement, the audience may still glean important cues to judge authorities. Such information may range from direct quotes and footage of authorities to journalists’ observations and interpretations about events involved in the decision-making process. Studies of mass communication have consistently found that use of news media, especially printed media such as newspapers or magazines, is a strong predictor of knowledge in public affairs (Andersen & Strömbäck, 2021; Hao et al., 2014; Robinson & Levy, 1996; Smith, 1986). In addition, community members paying more attention to local media tend to show more extreme attitudes toward local issues (Scheufele et al., 2002), greater interest, knowledge, and participation in local politics (McLeod et al., 1996), and increased use of multifaceted knowledge structures in voting decisions (Curnalia, 2010). Research has also found that media use facilitates elaboration about political figures (Birch & Allen, 2015), which may reinforce judgment certainty.

H2: Use of news media will be positively related to the certainty of trust judgments about authorities.

Third, community members and other stakeholders may discuss with each other the conduct and intentions of authorities. When facing important issues that could significantly impact the health or safety of their group, individuals may discuss the issue with others socially close to them, often engaging in persuasive attempts. Finding agreement in these discussions such as feedback that their attitudes are congruent with social consensus, can increase certainty of attitudes (Petrocelli et al., 2007; Visser & Mirabile, 2004). Interestingly, research has found that a hostile discussion environment can also reinforce attitude certainty either when one successfully resists a strong counterpersuasive appeal (Tormala & Petty, 2002, 2004) or fails to resist the appeal (Rucker & Petty, 2004). Valentino and Sears (1998) found that interpersonal communication about political events plays a pivotal role in political socialization processes in which political attitudes acquired during adolescence from close others such as family carry over into adulthood. These findings suggest that discussing the conduct or trustworthiness of an authority with others will increase one’s confidence in their judgments of the authority.

H3: Involvement in interpersonal discussion about authorities will be positively related to the certainty of trust judgments about authorities.

Finally, this study examines the extent to which these three communication antecedents explain unique or overlapping variance in attitude certainty. Considering the lack of theoretical guidance to formulate a hypothesis, the following research question is proffered:

RQ1: Do the three communication practices—direct engagement with authorities, use of news media, and involvement in interpersonal discussion—account for unique or overlapping variance in certainty of trust judgments about authorities?

Behavioral Intention as a Consequence of Trust Judgment Certainty

Research suggests that attitudes held with greater certainty are more likely to induce behaviors consistent with the attitude (Gross et al., 1995). In an early study, Fazio and Zanna (1978b) found that direct experience with an object increased attitude certainty, which in turn increased attitude-behavior consistency. Similarly, Berger and Mitchell (1989) found that participants who experienced the product directly or were exposed to an advertisement repeatedly for three to four times expressed greater confidence in their attitudes than those who were exposed to the ad only once. The former group was more consistent between their attitudes and choice behaviors than was the latter. A meta-analysis of 41 studies found that confidence in one’s attitude increased its stability over time which resulted in greater strength of the attitude-behavior correlation (Glasman & Albarracín, 2006). As a reasonable proxy for actual behavior (Fishbein & Ajzen, 1975; Kim & Hunter, 1993), this study focuses on behavioral intentions as the outcome of the interaction between trust judgments and certainty.

H4: The greater the certainty in trust judgments, the stronger will be the positive association between trust judgments and intentions to engage in trust-expressive behavior.

Marginal Value of Separately Measuring Certainty of Trust in Surveys

Finally, to justify the measurement of certainty which will require additional time and space in survey questionnaires, evidence must show that so doing will improve the explanatory power of models beyond what is afforded by asking about trust judgments alone. Thus, RQ2 asks how including certainty of trust improves predictions of behavioral intentions relative to a model using trust judgments as the sole independent variable.

RQ2: Does certainty of trust judgments account for variance in trust-expressive behavioral intentions in addition to that accounted for by trust judgments alone?

In sum, this paper seeks to address a gap in the trust and credibility literature which has fallen short of explaining variations in the association between trust and trust-expressive behavior. Inquiries in this paper test whether certainty of trust, shaped through communication processes, moderate the effects of trust on behavioral intentions. To this end, two empirical studies using minimal measurement of trust certainty were conducted, one based a local environmental context and another situated in a global health crisis.

Study 1

The first study used data collected from a survey in a local environmental dispute setting in which salt caverns under the scenic Seneca Lake in upstate New York were to be repurposed into liquefied petroleum gas (LPG) storages. This local context was chosen because the physical proximity of parties involved in the issue and the centrality of the issue to the residents’ interest enabled observation of various communication modes that were of interest to this study. At the time of data collection, the New York State Department of Environmental Conservation (NYSDEC) was reviewing the draft environmental impact statement submitted by a Texas-based energy company as part of the permit application process while various civic groups were organizing opposition movements.¹ Among the four counties having a shoreline at the lake (i.e., Schuyler, Yates, Ontario, Seneca), the legislature of Schuyler County which would host the gas storage was most supportive of the project, likely due to the prospect of increased job opportunities and tax revenues. In contrast, the other three counties adopted resolutions against the project, citing risks such as explosion and spillage during transportation and leakage of gas and saline from storages into the lake. The survey comprised two waves, both of which included measurements of attitude certainty and behavioral intentions. Data collection for the survey took place from August to October, 2017.

Method

Participants

Participants were residents of the four counties surrounding Seneca Lake. To facilitate contrasts based on their position toward the project, the study employed a stratified random sampling procedure to sample half of the participants from Schuyler County and the other half evenly from the three other counties. Using tax parcel data for the 2016 fiscal year made publicly available by the New York State (2017) government, 6,000 residential addresses were randomly selected (Schuyler: 3,000; other counties 1,000 each). Letters inviting participants to the survey were sent to the selected tax parcel addresses, explaining the purpose of the survey and offering participants two potential options to participate: online or mail. Out of 6,000 invitation mails sent, 1,215 were returned after failed delivery for various reasons including vacancy of the property, absence of mail receptacles, failure to forward, or invalid addresses. Of the 4,785 invitations that were not returned, 460 participants completed the first wave of the survey (381 online, 79 by mail) and 253 followed through to complete the second wave of the survey (195 online, 58 by mail). Thus, the response rate was 10% for the first wave, with a wave-to-wave retention rate of 55%. Demographically, 61% of the participants in Wave 1 were male. Participants’ age ranged from 19 to 94, with a median age of 63. In terms of education, 12% reported that they had received a high school diploma or less; 56% had some college education, an associate degree, or a bachelors’ degree; and 32% had a post-graduate degree. A vast majority (96%) identified themselves as white or Caucasian. Compared to American Community Survey data (US Census Bureau, 2017), males, elders, and those with higher educational attainment were somewhat overrepresented, likely due to the sampling method addressing invitations to property owners. Participants were very evenly distributed in terms of political party affiliation with 31% identifying as Democrat, 29% as Republican, and 30% as independent. As for political ideology, 46% identified themselves as at least leaning liberal, 38% as at least leaning conservative, and 15% as independent.

Procedure

Mailed letters invited participants to take part in the survey about the LPG storage project for a chance to enter a drawing for ten $100 gift cards redeemable at major retailers with information about the two participation methods. Participants who chose the online option were instructed to access the URL included in the letter to find the online survey and use a unique identification code to enter and complete it. For online participants, an email invitation to Wave 2 of the survey was sent in 3 weeks from their completion of Wave 1. Those who chose the mail option were instructed to either call or text the author with their identification code to request a printed copy of the survey, which was then delivered to them with a return envelope. The mail option was considered necessary because the study area was fairly rural with an expectedly considerable proportion of seniors in the population. For participants who did not respond to the survey and whose initial invitation mails were not returned in 2 weeks, a reminder postcard with participation instructions was delivered. A mail with Wave 2 of the survey and a return envelope was sent to mail participants in approximately 17 days from the receipt of their response to Wave 1. All responses returned via mail were entered into a spreadsheet which was later merged with data collected online.

Measurement and Instruments

Trust in NYSDEC

Participants rated their trust in NYSDEC in both Waves 1 and 2 using the same semantic differential scale. The scale consisted of five items adopted from the believability index developed by Meyer (1988) and successfully adapted in risk communication studies by McComas and Trumbo (2001) (Can be trusted/Can’t be trusted, Inaccurate/Accurate, Unfair/Fair, Tells the whole story/Doesn’t tell the whole story, Unbiased/Biased) and two additional word pairs that were deemed relevant to risk management (Not concerned about the public’s interest/Concerned about the public’s interest, Incompetent/Competent). Three of the seven items were reverse-coded. The scale was reliable in both waves, α_Wave1 = .94, α_Wave2 = .94.

Antecedents of Certainty of Trust Judgments

The three communication processes predicting the strength-related attributes of trust were measured in Wave 1. To measure direct engagement, the questionnaire asked how many times one had (1) spoken with, (2) seen, and (3) received a personalized email or letter from or had a telephone call with a NYSDEC employee in person over the last 5 years (1 = Not at all; 5 = A great deal). The scale was reliable, α = .84, and the mean of the three items was used for analysis.

To measure news media use, the survey employed two questions to ask the number of days in the previous week on which one watched or read any news covering local affairs (None to 7 days) and the level of attention paid to news covering local affairs (1 = None at all; 5 = A great deal) for three different types of news sources: television, print newspaper, and online news. The two items made up a reliable scale for all three types of media (Spearman-Brown coefficient for television: .80, newspaper: .81, online news: .79). After a mean index was calculated for each media type, the three values were summed into a single scale of news media use.

Four items measured engagement in interpersonal discussion by asking how often one engaged in the following activities: (1) discuss local politics with others (e.g., family friends, neighbors. . .), (2) discuss environmental issues in the local area with others, (3) discuss new development projects in the local area with others, and (4) attend public hearings or meetings in one’s community to discuss local affairs. All responses were recorded on a five-point scale (1 = Never; 5 = Always). Because the four items were reliable, α = .85, they were averaged into a composite scale of interpersonal discussion.

Certainty of Trust Judgments

Certainty of trust judgments was measured with a single-item question following the semantic-differential scale measuring trust using language adapted from standard attitude strength research (Wegener et al., 1995): “How certain do you feel about your ratings of NYSDEC you just provided?” (1 = Not at all certain; 5 = Extremely certain).

Intentions to Engage in Trust-Expressive Behavior

Finally, as the dependent variable, intentions to engage in behavior expressive of trust in NYSDEC (i.e., behavioral intentions) was measured with two items: “I will accept NYSDEC’s decision on this issue even if it is against my opinion,” and “I will support NYSDEC’s leadership in future environmental disputes regardless of their decision on the LPG storage case” (1 = Strongly disagree; 7 = Strongly agree). The two items comprised a reliable scale (Spearman Brown coefficients; Wave 1: .89, Wave 2: .90), and were averaged for analysis.

Data Analysis

Data were analyzed primarily using SPSS Statistics Version 25. Prior to all analyses, continuous independent variables were centered to the mean. The effects of communication antecedents of certainty were first tested individually and then jointly with and without control variables to determine if the variance explained by different antecedents overlap. To answer RQ2 which concerned whether certainty of trust accounts for unique variance in addition to trust, hierarchical regression models were used to assess if the R² value of the model increased when certainty and its interaction term with trust were added.

Results

On average, participants tended to oppose the LPG storage project. Responding to a question asking whether they supported or opposed the proposed project (1 = Strongly oppose; 7 = Strongly support), the participants’ median response was 2 with a mean of 2.83 (SD = 2.27). Schuyler County (M = 3.20, SD = 2.44) showed greater support for the project than did Yates (M = 2.01, SD = 1.76), Ontario (M = 2.43, SD = 1.83), and Seneca (M = 2.52, SD = 2.08), F (3, 454) = 6.63, p < .001, η² = .042. Overall, the valence of trust in NYSDEC (Wave 1) was slightly more positive (M = 4.42, SD = 1.44) than the scale midpoint (=4), t(456) = 6.22, p < .001, and the certainty of these judgments (M = 3.59, SD = .91) was above the scale midpoint (=3), t(456) = 13.99, p < .001.

H1 through H3 predicted that direct engagement with NYSDEC employees (H1), news media use (H2), and interpersonal discussion (H3) would be positively related to certainty of trust judgments. When the correlation of each of these predictors with certainty was tested separately, all three correlations were significant, direct engagement, r(455) = .29, p < .001, news media use, r(454) = .17, p < .001, involvement in interpersonal discussion, r(454) = .20, p < .001. Answering RQ1, all three relationships remained significant when they were simultaneously tested in a regression model controlling for each other (Table 1, Model 1), as well as when further control variables regarding support for the project and demographics were added to the model (Model 2).

Table 1.

Regression Results for Certainty of Trust Judgments With and Without Control Variables.

Variable	Model 1					Model 2
Variable	Β	SE Β	β	95% CI [LL, UL]	p Value	Β	SE Β	β	95% CI [LL, UL]	p Value
(Constant)	3.59	0.04		[3.51, 3.67]	.000	3.01	0.30		[2.42, 3.61]	.000
Direct Engagement	0.30	0.05	.26	[0.19, 0.40]	.000	0.25	0.06	.22	[0.14, 0.36]	.000
News media use	0.03	0.01	.13	[0.01, 0.05]	.007	0.02	0.01	.10	[0.00, 0.04]	.038
Interpersonal Discussion	0.11	0.05	.10	[0.00, 0.21]	.044	0.14	0.05	.12	[0.03, 0.24]	.012
Schuyler^a						−0.03	0.08	−.02	[−0.19, 0.13]	.725
Project support						0.09	0.02	.21	[0.05, 0.12]	.000
Sex^b						−0.12	0.09	−.06	[−0.29, 0.06]	.190
Age						0.01	0.00	.09	[0.00, 0.01]	.055
Race^c						0.26	0.22	.05	[−0.18, 0.70]	.242
Education^d						−0.03	0.08	−.01	[−0.19, 0.14]	.760
Political party^e						0.00	0.09	.00	[−0.19, 0.18]	.989
R ²	.12					0.17
Adj. R²	.11					0.15

Note. Β = unstandardized coefficient; β = standardized coefficient; CI = confidence interval; LL = lower limit; UL = upper limit.

0 = non-Schuyler County resident, 1 = Schuyler County resident.

0 = male, 1 = female.

0 = non-White, 1 = White.

0 = Without BA degree, 1 = BA or higher.

0 = non-Democrat, 1 = Democrat.

As a test for whether trust judgments of greater certainty would be more predictive of behavioral intentions, H4 predicted a positive interaction effect between trust judgments and the certainty of these judgments on trust-expressive behavioral intentions. In Wave 1, this effect was significant in the expected direction, F(1, 443) = 7.48, p = .006, $η_{p}^{2}$ = .017 (Figure 1a). In Wave 2, this positive interaction effect continued to remain significant, F(1, 252) = 6.14, p = .014, $η_{p}^{2}$ = .024 (Figure 1b). RQ2 asked if models including certainty perform better than models predicting behavioral intentions with trust but without certainty. In Wave 1, relative to the model without certainty of trust judgments, R² = .354, the model including certainty and its interaction term with trust, R² = .367, accounted for more variance in behavioral intentions, ΔR² = .013, p = .013. The same pattern emerged in Wave 2. The model with certainty R² = .378, accounted for more variance than did the model without certainty, R² = .348, ΔR² = .030, p = .003.

Figure 1.

The moderating effects of certainty on the relationship between trust and behavioral intentions. Panels (a), (b), and (c) represent findings from Waves 1 and 2 of Study 1 and Study 2, respectively.

As an alternative indicator of certainty effects, I also explored the persistence of behavioral intentions between the two waves. After coding the composite behavioral intentions index as either positive, neutral, or negative relative to the midpoint of the scale (=4), I coded persistence as 1 if the index stayed consistent in valence between the two waves, otherwise as 0. A binary logistic model with trust certainty in Wave 1 predicting behavioral intention persistence showed significant model fit, χ² (1, N = 238), 5.517, p = .019, Nagelkerke R² =.033. Certainty was positively associated with the persistence of behavioral intention, OR = 1.43, 95%CI [1.059, 1.939], further supporting how certainty affects trust-related consequences.

Study 2

Study 2 sought to address mainly two limitations of Study 1. First, Study 1’s findings were cross-sectional in nature, making it difficult to ascertain the causal direction of relationships. In addition, replication in a context beyond a local environmental dispute was necessary to enhance the external validity of the findings. As argued in the seminal work by Cook and Campbell (1979), generalizing research findings from a specific context requires replication in multiple heterogenous studies rather than a single large study. In contrast to the local context in Study 1 which afforded direct interactions and rich interpersonal cues for judging authorities, Study 2 used a national-level context in which the public rely mostly on mass-mediated cues. Using the COVID-19 pandemic as the setting, this online experiment manipulated message sources to test the role of certainty in participants’ intentions to comply with the source’s advice.

Among the three antecedents of trust judgment certainty tested in Study 1, Study 2 focused specifically on the effects of news media use due to the context situated at the national level. Participants were expected to receive information about the expert figures mostly through various forms of news media because it was very unlikely, for example, for most participants to have had any direct engagement with these national figures. In addition to assessing whether news use was positively associated with certainty (H2), the study also explored whether attention to the assigned figure in news mediated the effects of the expert figure manipulation on the level of certainty.

As for the consequences of trust judgment certainty, as in Study 1, Study 2 tested the moderating effects of certainty on trust judgments (H4). In addition, a mediation model was used to test whether the moderating effects of certainty was independent from that of trust when experimental manipulations cause variance in trust and certainty simultaneously.

Method

Participants

Participants were 703 adult Amazon Mechanical Turk workers in the United States. Despite mounting concerns over the quality of this sample pool (Cheung et al., 2017), researchers have also argued that such concerns could be successfully mitigated through proper screening (Chmielewski & Kucker, 2020; Kennedy et al., 2018). This study employed screening functions blocking low-quality participants built into CloudResearch, an MTurk extension platform (Litman, 2020). Participants were somewhat disproportionately male in gender (60%), and the median age was 34 years. About 75% reported having a college degree or higher and about 69% identified as middle class or higher. Politically, participants were slightly more liberal (M = 3.58, SD = 1.96) than the midpoint of the scale (1 = very liberal, 7 = very conservative).

Procedure and Instruments

The experiment was a between-subjects design with three experimental conditions. Participants were randomly assigned using the randomization function in Qualtrics. Comparison across three conditions revealed no difference in measured sociodemographic variables. Each condition featured one of the top governmental COVID-19 experts in the U.S. who were serving in the White House task force at the time: Drs. Anthony Fauci, Deborah Birx, and Robert Redfield. Because attitude strength typically takes time and repetition to build up in the real world, this approach intended to capitalize on the naturally occurring variance in trust judgment certainties across the three federal experts. Serving as the spokesperson of the White House Coronavirus Task Force’s daily briefings, Dr. Fauci became the face of the federal government’s early pandemic response, receiving considerable media attention in part because of his advice to the public which often contradicted President Trump’s. Dr. Birx took several questions during the press conferences alongside Dr. Fauci, but it was only after the data collection period (May 5–8, 2020) that she began playing the leading role as an expert figure in the press briefings. Dr. Redfield, Center for Disease Control director, also served in the White House Task Force but did not make regular media appearances as the other two. Confirming these assumptions, Google Trends data (Figure 2) shows that the three experts received clearly distinct levels of search interest over the period leading up to data collection.

Figure 2.

Google trends search interest results of the three expert figures used in experimental conditions of Study 2 between February 1, 2020 and May 4, 2020. Numbers represent search interest relative to the highest point (fixed as 100) on the chart for the given region and time.

Participants first rated their levels of trust, this time on a three-factor scale of perceived trustworthiness of governmental organizations developed by Grimmelikhuijsen and Meijer (2014). The scale consisted of 13 items measuring competence (α = .90), benevolence (α = .89), and honesty (α = .88). Because the three subscales were strongly correlated with each other (r_{competence-benevolence} = .869, r_{benevolence-honesty} = .869, r_{honesty-competence} = .906, all ps < .001) they were averaged into a single scale of trust. To assist participants’ evaluation, a photo of their judgment target was displayed above the question. In all three conditions, the photo featured the expert speaking at the podium in the White House briefing room with the U.S. flag and the White House seal in the background. The question asking the certainty of trust judgments similar to that used in Study 1 followed.

Measure of news use consisted of two items adjusted from Study 1: one item asked how much news covering COVID-19 one read or watched over the past 2 months and another asked how much attention they paid to such news (1 = none at all, 5 = a great deal). As in Study 1, the two items were averaged into a scale of general news use (α = .70). To additionally assess whether news use actually influenced differential levels of trust judgment certainty across expert figures, the survey also asked how much news featuring their assigned figure participants read or watched (1 = none at all, 5 = a great deal) and how prominently one thought the assigned figure was featured in news covering COVID-19 (1 = not prominently at all, 5 = extremely prominently). These two items were averaged into a scale of figure-specific news use (α = .80).

Behavioral intentions, the final dependent variable, included five items: openly supporting decisions made by the assigned figure affecting oneself personally, openly supporting decisions made by the assigned figure affecting the American public, following a new health practice recommended by the assigned figure, encouraging one’s family to follow a new health practice recommended by the assigned figure, and supporting budget proposals that will increase funding for policy initiatives recommended by the assigned figure (α = .94).

Results

Among the three sources, Dr. Fauci emerged as the most trusted across all three factors of competence, benevolence, and honesty (Figure 3). Participants were also most certain about the trust judgments they made about Dr. Fauci. Post-hoc comparisons with Bonferroni corrections indicated that for all three trust judgment factors, the differences between Fauci and Birx were significant (competence: p = .002, benevolence: p = .006, honesty: p = .001) but Birx and Redfield were not (competence: p = .144, benevolence: p = .250, honesty: p = 1.00). With judgment certainty, although the difference between the Fauci and Birx condition was not significant with Bonferroni corrections (p = .102), it was significant when adjustments were lifted (p = .034). I accordingly focus on the contrast between the Fauci condition versus the other two in the rest of the analysis.

Figure 3.

Level of trust toward each information source by factor (i.e., competence, benevolence, honesty) and as a composite index (measured on −2 to +2 scale) and certainty of judgments (measured on 1–5 scale). Error bars indicate 95% confidence interval for the mean.

H2 posited that the more people use news media, the more certain they will be about their trust judgments. General news use was positively associated with certainty of trust judgments, r = .32, p < .001. Furthermore, figure-specific news use mediated the relationship between the expert figure manipulation and trust judgment certainty (Figure 4). A mediation analysis using PROCESS Model 4 with 5,000 bootstraps revealed that relative to other conditions, participants in the Fauci condition used more news material featuring their message source, which in turn, led to higher levels of certainty of trust judgments, 95% CI of indirect effect: [0.170, 0.309]. Controlling for the indirect effects, the direct effect of the expert figure manipulation on certainty was not significant.

Figure 4.

The mediating effects of figure-specific news use in the effects of experimental condition on certainty of trust judgments. Coefficients in parentheses are standardized effects. Italicized coefficients indicate the total effect of the source manipulation certainty. Solid and broken arrows indicate significant and insignificant paths, respectively.

H4 held that trust judgment certainty would moderate the effect of trust on behavioral intentions such that the more certain one feels, the greater the positive association between trust and behavioral intentions will be. Supporting this hypothesis, the certainty × trust interaction was significant, F(1, 699) = 20.04, b = .182, p = .001, η_p = .028 (Figure 1c). Answering RQ2, it was examined whether the inclusion of certainty and its interaction term with trust would increase the amount of explained variance in behavioral intentions. Relative to the model using trust as the sole predictor of behavioral intentions, R² = .620, Adjusted-R² = .619, the model that also included certainty and the certainty × trust interaction term accounted for greater variance, R² = .637, Adjusted-R² = .636, F-change (2, 699) = 16.93, p < .001.

An additional analysis was conducted to further verify whether the interaction effects between trust and certainty accounted for the effect of the experimental condition on behavioral intentions, controlling for the indirect effect mediated by trust. A mediation model contrasting the Fauci condition against the other two was developed with the experimental condition as the independent variable, trust and certainty as two parallel mediators, behavioral intentions as the dependent variable, and the interaction term between the two mediators as a moderating variable predicting behavioral intentions. The moderated mediation model (Figure 5) shows that between the two indirect paths mediated by trust and certainty, only the indirect effect through trust was significant. Critically for this study, the interaction effect between trust and certainty on behavioral intentions was significant and positive, b = .18, SE = .05, p < .001, suggesting that trust judgments high in certainty were more likely to induce corresponding behavioral intentions. In sum, certainty further enhanced the effect of trust on behavioral intentions.

Figure 5.

Model showing moderation effects of certainty on the relationship between trust and behavioral intentions. Coefficients in parentheses are standardized effects. Solid arrows and broken arrows indicate significant and insignificant paths, respectively.

General Discussion

Despite the rich literature documenting antecedents or underlying dimensions of trust, few studies have sought to identify the conditions under which trust brings about the outcomes important to authorities. It is intuitively plausible that trust built upon extensive history of positive interactions is more likely to engender support and compliance with their decision than is more fleeting forms of “trust” based on vague expectations existing without recognition of specific authorities in charge. However, little research has directly addressed these important nuances across different states of trust. Findings reported in this paper suggest that research on attitude certainty can provide structural guidance predicting when trust will be effective, leading to enhanced behavioral conformity with authorities’ advice.

Even when participants indicate the same level of trust toward a message source, these judgments may still vary in terms of attitude strength attributes such as certainty. Findings from Study 1 show that communication processes such as direct engagement, media usage, and interpersonal discussion with others reinforced one’s certainty in trust judgments while controlling for the effects of each other, position on the environmental dispute, and demographic variables. This certainty, in turn, effectively predicted the extent to which trust judgments guide behavioral intentions as well as the stability of these intentions over time. Study 2 bolstered the findings of Study 1 regarding the role of trust judgment certainty using experimental evidence from another naturalistic yet broader context. As in Study 1, use of news media was a source of certainty when it comes to judgments about different expert figures addressing the COVID-19 pandemic. Furthermore, news media use specifically featuring the assigned figure mediated the relationship between the experimental manipulation and certainty. Study 2, as in Study 1, also found that while individuals were more inclined to follow the words of sources they trusted, this effect was further augmented when they felt more certain about their trust judgments. Although the experimental manipulation used here affected both trust and certainty, the effect of certainty on behavioral intentions functioned only as a moderator of trust.

For organizations engaging with the public, this suggests that they should consider not only whether the public judges them positively, but also how strong such judgments are. Building public trust involves not only eliciting momentary positive impressions, but also making efforts to solidify such impressions so that they become established as stable representations in memory. Findings here suggest that increasing communication experiences between the message source and the public can help achieve this goal. While directly engaging the audience will be particularly effective in increasing certainty, it appears that greater presence via news media as well as sparking discussion among the audience can also be effective measures. As for research, the implication is simple. Researchers should consider measuring attitude strength (e.g., certainty) whenever they use trust or credibility as a variable involving valenced judgments of an authority figure. This study shows that even a highly economical single item measure of certainty can increase the amount of explained variance in important consequences. It also suggests that trust and credibility research should more explicitly acknowledge the dual modes of trust judgments as thoroughly studied in general attitude research. Scholars should also use strength variables such as certainty to better learn how trust judgments vary in strength both as a result of communication practices and as an antecedent of cooperative behavior with authorities.

A few limitations of this paper must be noted. First, the current paper chose to focus on attitude certainty as a central attribute of attitude strength while neglecting the possible role of many others. For instance, the personal importance one attaches to an attitude (Howe & Krosnick, 2017), the accessibility of attitudes in memory (Fazio et al., 1982; Holland et al., 2003), and the ambivalence with which one holds an attitude (Thompson et al., 1995) have received considerable attention from researchers studying attitudes. In contexts of communication between authorities and the public, importance or accessibility related to individual or contextual factors may cause people to vary in their reliance on credibility or trust judgment cues. Similarly, people may hold ambivalent judgments between organizations (e.g., governmental agency) and individuals representing them (e.g., public official), a potential research area in which attitudinal ambivalence can bear significant implications.

Second, whereas the studies here focused only on the attitude-behavior link as the indicator of attitude strength, a more comprehensive analysis should also consider other strength-defining consequences including persistence, resistance, and impact on information processing. For instance, authorities could be particularly interested in understanding their own capacity to dispel and correct false beliefs while misinformation abounds (Scheufele & Krause, 2019) or choosing the right course of action after trust has been violated (Brühl et al., 2018). While the last analysis reported in Study 1 provides some preliminary insight into effects on persistence, future research can test more holistically how trust judgments interact with strength-related attributes of trust to impact outcomes beyond behavioral compliance.

Third, authority figure, which was the experimental manipulation in Study 2, affected both the proposed mediator (trust) and the moderator (certainty) in the tested model, partially compromising the interpretability of the interaction term. Because trust was conceptually considered as gaining strength through repeated communicative experiences, the choice of stimuli in this study was driven by the urge to use a natural source of variance in certainty with multiple risk managing authorities involved in a nationally visible context. Future studies seeking to verify the internal validity of the findings here may use manipulation isolating communication effects on certainty from those on trust in a lab environment.

Fourth, although operationalizations of trust in this study have followed the trust-as-attitude approach (Li, 2015) which views the trustor’s perception of the trustee’s trustworthiness as the essence of trust judgments, recent scholarship in trust research increasingly highlights the need to distinguish evaluations of trustworthiness from the trustor’s subsequent willingness to accept vulnerability (Hamm, 2017). Future applications of attitude strength to trust research may examine the specific role of attributes such as certainty in these more detailed conceptualizations of trust processes.

Finally, important moderators of trust and certainty warrant further research. Whereas this study tested the implications of certainty for trust judgments regarding risk-managing authorities, further research is needed to clarify whether it functions similarly in contexts where authorities are responsible for distributing benefits. In addition, in the increasingly polarized media environment, the political identity of a particular channel may be at odds with that of governmental authorities delivering public advice through that channel. Future research may explore how group identities of individuals and channels interact to influence trust and certainty.

To conclude, the present work underscores the importance of considering the strength-related dimensions of trust and credibility in both research and practice by showing that trust judgments held with greater certainty lead to greater intentions to cooperate with the judged authority. Findings reported in this paper demonstrate that direct and mediated communication practices between authorities and their constituents can increase the strength with which the latter judges the former. Strengthened judgments, in turn, will have greater impact on intentions to engage in trust-expressive behavior than will weaker judgments based on more superficial exposure to the authority. In addition to the extensive body of literature on how to improve one’s credibility and trustworthiness as a communicator, it appears also very important to complement this knowledge with further research on how authorities might bolster the audience’s certainty, as well as other forms of strength, in these judgments.

Footnotes

Acknowledgements

The author thanks Katherine A. McComas for her inspiration and thoughtful advice while carrying out Study 1 of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by the National Science Foundation (#1658075) and the Brian Lamb School of Communication.

ORCID iD

Hwanseok Song

Notes

Author Biography

Hwanseok Song is an assistant professor of communication at the Brian Lamb School of Communication at Purdue University. His primary research interests focus on how people make and use social judgments in contexts where they communicate information about science, safety, environmental issues, and technological risks.

References

Allum

(2007). An empirical test of competing theories of hazard-related trust: The case of GM food. Risk Analysis, 27(4), 935–946. https://doi.org/10.1111/j.1539-6924.2007.00933.x

Andersen

Strömbäck

(2021). Media platforms and political learning: The democratic challenge of news consumption on computers and mobile devices. International Journal of Communication, 15, 20.

Baek

Y. M.

Wojcieszak

Delli Carpini

M. X.

(2012). Online versus face-to-face deliberation: Who? Why? What? With what effects? New Media & Society, 14(3), 363–383. https://doi.org/10.1177/1461444811413191

Berger

I. E.

Mitchell

A. A.

(1989). The effect of advertising on attitude accessibility, attitude confidence, and the attitude-behavior relationship. Journal of Consumer Research, 16(3), 269–279.

Berlo

D. K.

Lemert

J. B.

Mertz

R. J.

(1969). Dimensions for evaluating the acceptability of message sources. Public Opinion Quarterly, 33(4), 563–576. https://doi.org/10.1086/267745

Birch

Allen

(2015). Judging politicians: The role of political attentiveness in shaping how people evaluate the ethical behaviour of their leaders. European Journal of Political Research, 54(1), 43–60. https://doi.org/10.1111/1475-6765.12066

Bohner

Dickel

(2010). Attitudes and attitude change. Annual Review of Psychology, 62(1), 391–417. https://doi.org/10.1146/annurev.psych.121208.131609

Brühl

Basel

J. S.

Kury

M. F.

(2018). Communication after an integrity-based trust violation: How organizational account giving affects trust. European Management Journal, 36(2), 161–170. https://doi.org/10.1016/j.emj.2017.08.001

Chaiken

(1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39(5), 752–766. https://doi.org/10.1037/0022-3514.39.5.752

10.

Cheung

J. H.

Burns

D. K.

Sinclair

R. R.

Sliter

(2017). Amazon mechanical Turk in organizational psychology: An evaluation and practical recommendations. Journal of Business and Psychology, 32(4), 347–361. https://doi.org/10.1007/s10869-016-9458-5

11.

Chmielewski

Kucker

S. C.

(2020). An MTurk crisis? Shifts in data quality and the impact on study results. Social Psychological and Personality Science, 11(4), 464–473. https://doi.org/10.1177/1948550619875149

12.

Cho

Shah

D. V.

McLeod

J. M.

McLeod

D. M.

Scholl

R. M.

Gotlieb

M. R.

(2009). Campaigns, reflection, and deliberation: Advancing an O-S-R-O-R model of communication effects. Communication Theory, 19(1), 66–88. https://doi.org/10.1111/j.1468-2885.2008.01333.x

13.

Conrey

F. R.

Smith

E. R.

(2007). Attitude representation: Attitudes as patterns in a distributed, connectionist representational system. Social Cognition, 25(5), 718–735. https://doi.org/10.1521/soco.2007.25.5.718

14.

Cook

T. D.

Campbell

D. T

. (1979). Quasi-experimentation: Design & analysis issues for field settings. Houghton Mifflin.

15.

Curnalia

R. M. L.

(2010). Predictors of the development of applied knowledge structures for presidential candidates. Communication Research Reports, 27(1), 80–89. https://doi.org/10.1080/08824090903526596

16.

Cvetkovich

Siegrist

Murray

Tragesser

(2002). New information and social trust: Asymmetry and perseverance of attributions about hazard managers. Risk Analysis, 22(2), 359–367. https://doi.org/10.1111/0272-4332.00030

17.

Dunn

J. R.

Schweitzer

M. E.

(2005). Feeling and believing: The influence of emotion on trust. Journal of Personality and Social Psychology, 88(5), 736–748. https://doi.org/10.1037/0022-3514.88.5.736

18.

Eagly

A. H.

Chaiken

(1993). The psychology of attitudes. Harcourt, Brace, & Jovanovich.

19.

Earle

T. C.

(2010). Trust in risk management: A model-based review of empirical research. Risk Analysis, 30(4), 541–574. https://doi.org/10.1111/j.1539-6924.2010.01398.x

20.

Earle

T. C.

Cvetkovich

(1995). Social trust: Toward a cosmopolitan society. Greenwood Publishing Group.

21.

Earle

T. C.

Siegrist

Gutscher

(2007). Trust, risk perception, and the TCC model of cooperation. In Siegrist

Earle

T. C.

Gutscher

(Eds.), Trust in cooperative risk management: Uncertainty and scepticism in the public mind (pp. 1–50). Earthscan.

22.

Fazio

R. H.

(2007). Attitudes as object–evaluation associations of varying strength. Social Cognition, 25(5), 603–637. https://doi.org/10.1521/soco.2007.25.5.603

23.

Fazio

R. H.

Chen

McDonel

E. C.

Sherman

S. J.

(1982). Attitude accessibility, attitude-behavior consistency, and the strength of the object-evaluation association. Journal of Experimental Social Psychology, 18(4), 339–357. https://doi.org/10.1016/0022-1031(82)90058-0

24.

Fazio

R. H.

Zanna

M. P.

(1978a). Attitudinal qualities relating to the strength of the attitude-behavior relationship. Journal of Experimental Social Psychology, 14(4), 398–408. https://doi.org/10.1016/0022-1031(78)90035-5

25.

Fazio

R. H.

Zanna

M. P.

(1978b). Attitudinal qualities relating to the strength of the attitude-behavior relationship. Journal of Experimental Social Psychology, 14(4), 398–408. https://doi.org/10.1016/0022-1031(78)90035-5

26.

Fazio

R. H.

Zanna

M. P.

(1981). Direct experience and attitude-behavior consistency. In Berkowitz

(Ed.), Advances in experimental social psychology (Vol. 14, pp. 161–202). Academic Press. https://doi.org/10.1016/S0065-2601(08)60372-X

27.

Fishbein

Ajzen

(1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Addison-Wesley PubCo.

28.

Frewer

L. J.

Howard

Hedderley

Shepherd

(1996). What determines trust in information about food-related risks? Underlying psychological constructs. Risk Analysis, 16(4), 473–486. https://doi.org/10.1111/j.1539-6924.1996.tb01094.x

29.

Fullwood

(2007). The effect of mediation on impression formation: A comparison of face-to-face and video-mediated conditions. Applied Ergonomics, 38(3), 267–273. https://doi.org/10.1016/j.apergo.2006.06.002

30.

Gawronski

Bodenhausen

G. V.

(2007). Unraveling the processes underlying evaluation: Attitudes from the perspective of the APE model. Social Cognition, 25(5), 687–717. https://doi.org/10.1521/soco.2007.25.5.687

31.

Glasman

L. R.

Albarracín

(2006). Forming attitudes that predict future behavior: A meta-analysis of the attitude-behavior relation. Psychological Bulletin, 132(5), 778–822. https://doi.org/10.1037/0033-2909.132.5.778

32.

Grimmelikhuijsen

S. G.

Meijer

A. J.

(2014). Effects of transparency on the perceived trustworthiness of a government organization: Evidence from an online experiment. Journal of Public Administration Research and Theory, 24(1), 137–157. https://doi.org/10.1093/jopart/mus048

33.

Gross

S. R.

Holtz

Miller

(1995). Attitude certainty. In Attitude strength: Antecedents and consequences (Vol. 4, pp. 215–245). Lawrence Erlbaum Associates.

34.

Hamm

J. A.

(2017). Trust, trustworthiness, and motivation in the natural resource management context. Society & Natural Resources, 30(8), 919–933. https://doi.org/10.1080/08941920.2016.1273419

35.

Hao

Wen

George

(2014). News consumption and political and civic engagement among young people. Journal of Youth Studies, 17(9), 1221–1238. https://doi.org/10.1080/13676261.2014.901490

36.

Holland

R. W.

Verplanken

van Knippenberg

(2003). From repetition to conviction: Attitude accessibility as a determinant of attitude certainty. Journal of Experimental Social Psychology, 39(6), 594–601. https://doi.org/10.1016/S0022-1031(03)00038-6

37.

Hovland

C. I.

Janis

I. L.

Kelley

H. H.

(1953). Communication and persuasion: Psychological studies of opinion change. Yale University Press.

38.

Howe

L. C.

Krosnick

J. A.

(2017). Attitude strength. Annual Review of Psychology, 68(1), 327–351. https://doi.org/10.1146/annurev-psych-122414-033600

39.

Hunt

Frewer

L. J.

Shepherd

(1999). Public trust in sources of information about radiation risks in the UK. Journal of Risk Research, 2(2), 167–180. https://doi.org/10.1080/136698799376916

40.

Isen

A. M.

Shalker

T. E.

Clark

Karp

(1978). Affect, accessibility of material in memory, and behavior: A cognitive loop? Journal of Personality and Social Psychology, 36(1), 1–12. https://doi.org/10.1037/0022-3514.36.1.1

41.

Johnson

B. B.

(1999). Exploring dimensionality in the origins of hazard-related trust. Journal of Risk Research, 2(4), 325–354. https://doi.org/10.1080/136698799376763

42.

Kennedy

Clifford

Burleigh

Jewell

Waggoner

(2018). The shape of and solutions to the MTurk quality crisis (SSRN Scholarly Paper No. ID 3272468). Social Science Research Network. https://papers.ssrn.com/abstract=3272468

43.

Kim

M.-S.

Hunter

J. E.

(1993). Relationships among attitudes, behavioral intentions, and behavior: A meta-analysis of past research, part 2. Communication Research, 20(3), 331–364. https://doi.org/10.1177/009365093020003001

44.

Krosnick

J. A.

(1988). The role of attitude importance in social evaluation: A study of policy preferences, presidential candidate evaluations, and voting behavior. Journal of Personality and Social Psychology, 55(2), 196–210. https://doi.org/10.1037/0022-3514.55.2.196

45.

Krosnick

J. A.

Petty

R. E.

(1995). Attitude strength: An overview. In Petty

R. E.

Krosnick

J. A.

(Eds.), Attitude strength: Antecedents and consequences (pp. 1–24). Lawrence Erlbaum Associates.

46.

P. P.

(2007). Towards an interdisciplinary conceptualization of trust: A typological approach. Management and Organization Review, 3(3), 421–445. https://doi.org/10.1111/j.1740-8784.2007.00081.x

47.

P. P.

(2015). Trust as a leap of hope for transaction value: A two-way street above and beyond trust propensity and expected trustworthiness. In Bornstein

B. H.

Tomkins

A. J.

(Eds.), Motivating cooperation and compliance with authority: The role of institutional trust (pp. 37–53). Springer International Publishing. https://doi.org/10.1007/978-3-319-16151-8_3

48.

Litman

(2020, July 15). New solutions dramatically improve research data quality on MTurk. CloudResearch. Retrieved January 31, 2022, from https://www.cloudresearch.com/resources/blog/new-tools-improve-research-data-quality-mturk/

49.

McComas

K. A.

Trumbo

C. W.

(2001). Source credibility in environmental health – risk controversies: Application of Meyer’s credibility index. Risk Analysis, 21(3), 467–480. https://doi.org/10.1111/0272-4332.213126

50.

McGinnies

Ward

C. D.

(1980). Better liked than right: Trustworthiness and expertise as factors in credibility. Personality and Social Psychology Bulletin, 6(3), 467–472. https://doi.org/10.1177/014616728063023

51.

McLeod

J. M.

Daily

Guo

Eveland

W. P.

Bayer

Yang

Wang

(1996). Community integration, local media use, and democratic processes. Communication Research, 23(2), 179–209. https://doi.org/10.1177/009365096023002002

52.

Metzger

M. J.

(2007). Making sense of credibility on the Web: Models for evaluating online information and recommendations for future research. Journal of the American Society for Information Science and Technology, 58(13), 2078–2091. https://doi.org/10.1002/asi.20672

53.

Metzger

M. J.

Flanagin

A. J.

Medders

R. B.

(2010). Social and heuristic approaches to credibility evaluation online. Journal of Communication, 60(3), 413–439. https://doi.org/10.1111/j.1460-2466.2010.01488.x

54.

Meyer

(1988). Defining and measuring credibility of newspapers: Developing an index. Journalism Quarterly, 65(3), 567–574. https://doi.org/10.1177/107769908806500301

55.

Nabatchi

Amsler

L. B.

(2014). Direct public engagement in local government. The American Review of Public Administration, 44(4_suppl), 63S–88S. https://doi.org/10.1177/0275074013519702

56.

Nayakankuppam

Priester

J. R.

Kwon

J. H.

Donovan

L. A. N.

Petty

R. E.

(2018). Construction and retrieval of evaluative judgments: The attitude strength moderation model. Journal of Experimental Social Psychology, 76, 54–66. https://doi.org/10.1016/j.jesp.2017.12.005

57.

New York State. (2017). NYS GIS clearinghouse. Retrieved May 3, 2018, from https://gis.ny.gov/

58.

O’Keefe

D. J.

(2015). Persuasion: Theory and research (3rd ed.). SAGE.

59.

Olson

G. M.

Olson

J. S.

(2000). Distance matters. Human–Computer Interaction, 15(2–3), 139–178. https://doi.org/10.1207/S15327051HCI1523_4

60.

Patzer

G. L.

(1983). Source credibility as a function of communicator physical attractiveness. Journal of Business Research, 11(2), 229–241. https://doi.org/10.1016/0148-2963(83)90030-9

61.

Peters

R. G.

Covello

V. T.

McCallum

D. B.

(1997). The determinants of trust and credibility in environmental risk communication: An empirical study. Risk Analysis, 17(1), 43–54. https://doi.org/10.1111/j.1539-6924.1997.tb00842.x

62.

Petrocelli

J. V.

Tormala

Z. L.

Rucker

D. D.

(2007). Unpacking attitude certainty: Attitude clarity and attitude correctness. Journal of Personality and Social Psychology, 92(1), 30–41. https://doi.org/10.1037/0022-3514.92.1.30

63.

Petty

R. E.

Briñol

Tormala

Z. L.

(2007). The role of metacognition in social judgment. In Higgins

E. T.

Kruglanski

A. W.

(Eds.), Social psychology: Handbook of basic principles (2nd ed., pp. 254–284). Guilford Press.

64.

Petty

R. E.

Cacioppo

J. T.

(1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123–205.

65.

Petty

R. E.

Haugtvedt

C. P.

Smith

S. M.

(1995). Elaboration as a determinant of attitude strength: Creating attitudes that are persistent, resistant, and predictive of behavior. In Petty

R. E.

Krosnick

J. A.

(Eds.), Attitude strength: Antecedents and consequences (Vol. 4, pp. 93–130). Lawrence Erlbaum Associates.

66.

Petty

R. E.

Krosnick

J. A.

(1995). Attitude strength: Antecedents and consequences. Psychology Press.

67.

Pornpitakpan

(2004). The persuasiveness of source credibility: A critical review of five decades’ evidence. Journal of Applied Social Psychology, 34(2), 243–281. https://doi.org/10.1111/j.1559-1816.2004.tb02547.x

68.

Robinson

J. P.

Levy

M. R.

(1996). News media use and the informed public: A 1990s update. Journal of Communication, 46(2), 129–135. https://doi.org/10.1111/j.1460-2466.1996.tb01478.x

69.

Rousseau

D. M.

Sitkin

S. B.

Burt

R. S.

Camerer

(1998). Not so different after all: A cross-discipline view of trust. Academy of Management. The Academy of Management Review, 23(3), 393–404.

70.

Rucker

D. D.

Petty

R. E.

(2004). When resistance is futile: Consequences of failed counterarguing for attitude certainty. Journal of Personality and Social Psychology, 86(2), 219–235. https://doi.org/10.1037/0022-3514.86.2.219

71.

Rucker

D. D.

Petty

R. E.

Briñol

(2008). What’s in a frame anyway? A meta-cognitive analysis of the impact of one versus two sided message framing on attitude certainty. Journal of Consumer Psychology, 18(2), 137–149. https://doi.org/10.1016/j.jcps.2008.01.008

72.

Scheufele

D. A.

Krause

N. M.

(2019). Science audiences, misinformation, and fake news. Proceedings of the National Academy of Sciences, 116(16), 7662–7669. https://doi.org/10.1073/pnas.1805871115

73.

Scheufele

D. A.

Shanahan

Kim

S.-H.

(2002). Who cares about local politics? Media influences on local political involvement, issue awareness, and attitude strength. Journalism & Mass Communication Quarterly, 79(2), 427–444. https://doi.org/10.1177/107769900207900211

74.

Schuldt

J. P.

Konrath

S. H.

Schwarz

(2011). “Global warming” or “climate change”? Whether the planet is warming depends on question wording. Public Opinion Quarterly, 75(1), 115–124. https://doi.org/10.1093/poq/nfq073

75.

Schuman

Presser

Ludwig

(1981). Context effects on survey responses to questions about abortion. Public Opinion Quarterly, 45(2), 216–223. https://doi.org/10.1086/268652

76.

Schwarz

(1999). Self reports: How the questions shape the answers. American Psychologist, 54(2), 93–105. https://doi.org/10.1037/0003-066X.54.2.93

77.

Schwarz

(2007). Attitude construction: Evaluation in context. Social Cognition, 25(5), 638–656. https://doi.org/10.1521/soco.2007.25.5.638

78.

Schwarz

Strack

Kommer

Wagner

(1987). Soccer, rooms, and the quality of your life: Mood effects on judgments of satisfaction with life in general and with specific domains. European Journal of Social Psychology, 17(1), 69–79. https://doi.org/10.1002/ejsp.2420170107

79.

Shah

D. V.

McLeod

D. M.

Rojas

Cho

Wagner

M. W.

Friedland

L. A.

(2017). Revising the communication mediation model for a new political communication ecology. Human Communication Research, 43(4), 491–504. https://doi.org/10.1111/hcre.12115

80.

Siegrist

(2021). Trust and risk perception: A critical review of the literature. Risk Analysis, 41(3), 480–490. https://doi.org/10.1111/risa.13325

81.

Siegrist

Connor

Keller

(2012). Trust, confidence, procedural fairness, outcome fairness, moral conviction, and the acceptance of GM field experiments. Risk Analysis, 32(8), 1394–1403. https://doi.org/10.1111/j.1539-6924.2011.01739.x

82.

Siegrist

Cvetkovich

Roth

(2000). Salient value similarity, social trust, and risk/benefit perception. Risk Analysis, 20(3), 353–362. https://doi.org/10.1111/0272-4332.203034

83.

Smith

H. H.

(1986). Newspaper readership as a determinant of political knowledge and activity. Newspaper Research Journal, 7(2), 47–54.

84.

Song

. (2020). Perceived standing: Exploring why people accept or reject others’ access to public participation in local environmental conflicts. Local Environment, 25(2), 397–413. https://doi.org/10.1080/13549839.2020.1753667

85.

Tanis

Postmes

(2003). Social cues and impression formation in CMC. Journal of Communication, 53(4), 676–693. https://doi.org/10.1111/j.1460-2466.2003.tb02917.x

86.

Thompson

M.M.

Zanna

M. P.

Griffin

D. W

. (1995). Let’s not be indifferent about (attitudinal) ambivalence. In Petty

R. E.

Krosnick

J. A.

(Eds.), Attitude strength: Antecedents and consequences (pp. 361–386). Lawrence Erlbaum Associates.

87.

Tormala

Z. L.

Petty

R. E.

(2002). What doesn’t kill me makes me stronger: The effects of resisting persuasion on attitude certainty. Journal of Personality and Social Psychology, 83(6), 1298–1313. https://doi.org/10.1037/0022-3514.83.6.1298

88.

Tormala

Z. L.

Petty

R. E.

(2004). Source credibility and attitude certainty: A metacognitive analysis of resistance to persuasion. Journal of Consumer Psychology, 14(4), 427–442. https://doi.org/10.1207/s15327663jcp1404_11

89.

Tormala

Z. L.

Rucker

D. D.

(2007). Attitude certainty: A review of past findings and emerging perspectives. Social and Personality Psychology Compass, 1(1), 469–492. https://doi.org/10.1111/j.1751-9004.2007.00025.x

90.

Trumbo

C. W.

(1999). Heuristic-systematic information processing and risk judgment. Risk Analysis, 19(3), 391–400. https://doi.org/10.1111/j.1539-6924.1999.tb00415.x

91.

US Census Bureau. (2017). American Community Survey (ACS). Author. Retrieved December 30, 2019, from https://www.census.gov/programs-surveys/acs

92.

Valentino

N. A.

Sears

D. O

. (1988). Event-driven political communication and the preadult socialization of partisanship. Political Behavior, 20(2), 127–154. https://doi.org/10.1023/A:1024880713245

93.

Visser

P. S.

Mirabile

R. R.

(2004). Attitudes in the social context: The impact of social network composition on individual-level attitude strength. Journal of Personality and Social Psychology, 87(6), 779–795. https://doi.org/10.1037/0022-3514.87.6.779

94.

Walls

Pidgeon

Weyman

Horlick-Jones

(2004). Critical trust: Understanding lay perceptions of health and safety risk regulation. Health, Risk & Society, 6(2), 133–150. https://doi.org/10.1080/1369857042000219788

95.

Wegener

D. T.

Downing

Krosnick

J. A.

Petty

R. E.

(1995). Measures and manipulations of strength-related properties of attitudes: Current practice and future directions. In Petty

R. E.

Krosnick

J. A.

(Eds.), Attitude strength: Antecedents and consequences (pp. 455–487). Lawrence Erlbaum Associates.

96.

Whitehead

J. L.

(1968). Factors of source credibility. Quarterly Journal of Speech, 54(1), 59–63. https://doi.org/10.1080/00335636809382870

97.

Wood

Rhodes

Biek

(1995). Working knowledge and attitude strength: An information-processing analysis. In Petty

R. E.

Krosnick

J. A.

(Eds.), Attitude strength: Antecedents and consequences (pp. 189–202). Lawrence Erlbaum Associates.

98.

Zuwerink

J. R.

Devine

P. G.

(1996). Attitude importance and resistance to persuasion: It’s not just the thought that counts. Journal of Personality and Social Psychology, 70(5), 931–944. https://doi.org/10.1037/0022-3514.70.5.931