Are Some People Less Influenced by Others' Opinions? The Role of Internal Political Self-Efficacy and Need for Cognition in Impression Formation on Social Networking Sites

Abstract

While social information processing theory (SIP) asserts that Internet users form impressions of others by using available cues in the web environment, questions remain regarding individual differences in dependence on such cues. Focusing on others' opinion cues, this study investigates the role of internal political self-efficacy (IPSE) and need for cognition (NFC) as potential moderators of the cue effect. Participants (N=251) were randomly exposed to a fictitious political candidate's Facebook profile displaying either positive or negative comments and then asked to rate perceived trustworthiness and expertise of the candidate. The results indicated that although others' opinions were powerful cues, their influence was moderated by IPSE; individuals with higher IPSE were less influenced by others' opinions. The two-way interaction was significantly moderated by NFC such that those with low IPSE and low NFC were most susceptible to others' influences, thereby indicating a significant three-way interaction among IPSE, NFC, and others' opinion cues.

Introduction

Social networking sites (SNS) are one of the primary avenues through which today's young adults access political news and find opportunities for participation.¹ Instead of reading a newspaper or watching the evening news, they open up a SNS page and become exposed to political posts and comments.² As SNS penetrate every area of social life, gaining support on SNS has become crucial for political candidates in order to win an election.^3,4 SNS are now a must-do or use-it-or-lose-it campaign tool for candidates.⁴

One of the primary features of SNS is others' contributions to one's profile (e.g., posts, comments). On SNS such as Facebook and MySpace, a profile displays not only the profile owner's posts but also others' posts or comments underneath each post. These posts or comments have the same or similar visibility. This structure of SNS has enabled third-party opinions to have more weight than in other communication situations, including traditional Web sites where readers' comments are hardly displayed in parallel with the content the Web site owner presents. Studies have demonstrated that information provided by others on SNS is perceived to be genuine and, therefore, affects observers' impressions of the profile owner more strongly than his or her self-presentation does.^5,6 These findings indicate the powerful influence of others' comments in current media environments. However, questions remain unanswered regarding what factors make individuals more or less vulnerable to the influence.

Guided by social information processing theory (SIP), the present study explores individual differences in reliance on others' opinion cues. By testing the roles of internal political self-efficacy (IPSE) and need for cognition (NFC), in particular, this study aims to offer psychological explanations for why some SNS users are more influenced by others' comments than others when they form an impression of an unknown target person. Previous studies examined IPSE and NFC mostly in relation to individuals' political understanding and participations, but little is known about their effects at the impression or perception level. By investigating the relationship among IPSE, NFC, and others' opinion cues, this study attempts to provide a better understanding of differences in susceptibility to others' opinion cues depending on their individual characteristics.

Others' opinion cues in impression formation through computer-mediated communication

Heuristics are learned knowledge structure or cognitive shortcuts such as source credibility, number of argument, and others' reactions.⁷ As cognitive misers, humans wish to satisfy their goal-related needs efficiently with the least effort.⁸ When they are highly motivated to make an accurate judgment (e.g., when the judgment is important to them), they are willing to expend cognitive effort in collecting and comprehending substantial information to reach a sufficient degree of confidence.⁵ When their motivation level is low (e.g., when the decision is not important) or when relevant substantial information is not accessible, they tend toward heuristic-based, less effortful decision-making processes.^9,10

Heuristics are particularly important in computer-mediated communication (CMC) where nonverbal, social context cues are lacking. SIP theory claims that by utilizing available cues in the online environment, Internet users form impressions of distant others with a similar level of accuracy to that of face-to-face communicators.¹¹ The cues include others' opinions about the target person; in impression formation, those “others” around the target person function as a context that attributes certain characteristics to the person¹² and significantly affects individuals' impressions regardless of the representativeness of the opinions.^13,14 Individuals tend to regard a few others' visible reactions as a snapshot of the typical or predominant opinion about the person.¹⁵

Impression can be conceptualized as credibility perception,¹⁵ which has two key components: trustworthiness and expertise.^16,17 When people talk about their impression of someone, they tend to mention how trustworthy the person seems to be (personal aspect) or how much expertise the person seems to have (professional aspect). Trustworthiness refers to the perceived honesty, character, and safety of the communicator.¹⁸ Expertise refers to the ability to perform qualitatively well in a particular task domain.¹⁹ The trustworthiness and expertise of a message sender are significantly related to the message receiver's attitude change and behavioral compliance.^20–22 Given that personal impressions of an unknown person are often influenced by how others talk about the person (i.e., others' opinion cues),⁵ it stands to reason that these core dimensions of impression—perceived trustworthiness and expertise—are also likely to be susceptible to others' opinion cues. Thus, the following hypothesis is posited:

H1: Those who are exposed to positive (vs. negative) comments on a target person are more likely to perceive the person as (a) trustworthy and (b) having expertise.

The role of IPSE and NFC

Individuals process information differently depending on their psychological and motivational needs.²³ One psychological factor that seems to moderate others' influences is IPSE, which is defined as “individuals' beliefs about their own competence to understand, and to participate effectively in politics.”²⁴ High-IPSE people tend to be more interested in politics and more active in political participation or information seeking.^25,26 On the other hand, low-IPSE people are less confident in their own capabilities to understand and judge political issues and less motivated to participate independently in politics, and therefore, they are more likely to be vulnerable to external influences. For instance, Holbert et al.²⁷ found that low-IPSE people were significantly more influenced by Jon Stewart's satirical, negative commentary than high-IPSE people. To make a political judgment, low-IPSE people may refer to others' opinion cues more heavily, whereas high-IPSE people are less dependent on what others have to say. Thus, the following hypothesis is posited:

H2: People with higher IPSE are less likely to be influenced by others' comments in perceiving (a) trustworthiness and (b) expertise of a target person.

Another individual factor that may moderate the effects of others' opinion cues is NFC. NFC is defined as an “individual's need to organize his experience meaningfully” and “understand and make reasonable the experiential world.”^28(p291) It involves an individual's “tendency to engage in and enjoy effortful cognitive activity.”^29(p198) Although correlated with intelligence, NFC is not a measure of intelligence but a measure of cognitive motivation.¹² Individuals who score high in NFC prefer employing effortful thinking to accomplish a goal and enjoy a task that involves problem solving, whereas those low in NFC only think as hard as they have to.³⁰ Because individuals with high NFC have high standards for cognitive clarity, they think harder than low-NFC individuals before they make a judgment. High-NFC people would make judgments based on a running tally of information and would keep ambivalent attitudes rather than rushing to a hasty conclusion based on a few superficial cues.³¹ They behave in line with the rational model in which individuals base their judgments on the quality of factual information and arguments.³¹ Considering that user comments available on SNS at the moment of exposure cannot be a representative sample of public opinion, high-NFC people are likely to put less weight on the comments in their evaluations. In contrast, low-NFC people may jump to conclusion based on the comments as they have little desire to assess the validity of cues. Thus, the following hypothesis is posited:

H3: People with high NFC are less likely to be influenced by others' comments in perceiving (a) trustworthiness and (b) expertise of a target person.

If IPSE and NFC individually moderate the effect of others' opinion cues, how do IPSE and NFC interact? For instance, is the vulnerability to others' influences even higher when both IPSE and NFC are low? The question raises the following research question:

RQ1: How do IPSE, NFC, and others' comments interact in perceiving (a) trustworthiness or (b) expertise of a target person?

Method

Participants and procedures

Participants were 251 college students recruited from a large university for extra credit (age: M=20.94 years, SD=1.99 years; female=52.19%; white=74.10%). Participants were randomly assigned into one of the two comment conditions (positive vs. negative) and saw a screenshot of a Facebook profile of a fictitious political candidate before answering a questionnaire.

Measures

Others' opinions were manipulated through the profile that displayed two comments: one posted by a man and one by a woman (see Figures 1 and 2 for examples). For better generalizability, four different people's profile photos were used for the candidate and the commenters in each condition. Also, both commenters' ages were manipulated to be either young (in their late teens or early 20s) or older (over 60s).

FIG. 1.

An example of the positive comments version (with the young commenters' photos among multiple stimuli). The photos are blurred for publication purposes.

FIG. 2.

An example of the negative comments version (with the older commenters' photos among multiple stimuli). The photos are blurred for publication purposes.

Perceived trustworthiness and expertise were measured using Ohanian's credibility scale that is composed of ten pairs of adjectives (e.g., trustworthy-untrustworthy and reliable-unreliable for trustworthiness, Cronbach's α=0.98; expert-not an expert, knowledgeable-unknowledgeable for expertise, Cronbach's α=0.95).¹⁷ Other variables were measured on an 11-point scale ranging from 0 to 10, and divided by 10. Candidates were rated on average to be medium trustworthy (M=0.50, SD=0.25), and a little higher than the midpoint in expertise (M=0.56, SD=0.20). IPSE was measured with two questions²⁷ (“Sometimes politics seems so complicated that a person like me cannot really understand what is going on” and “People like me have no say over what the government does,” both reverse-coded; Pearson r=0.22, p<0.001; M=0.65, SD=0.18). NFC was measured with 18 questions²⁵ (e.g., “I would prefer complex to simple problems,” “It's enough for me that something gets the job done; I don't care how or why it works [reverse-coded]”; Cronbach's α=0.63; M=0.57, SD=0.13). Participants' sex; race (white vs. nonwhite); frequency of Facebook use (“never” to “all the time”; M=0.74, SD=0.27); ideology (an index of three items about social, economic, and security issues: “very liberal” to “very conservative”; Cronbach's α=0.87; M=0.52, SD=0.28); political interest (how closely they followed political news: “not at all closely” to “very closely”; M=0.63, SD=0.22); general trust in politicians (“Politicians are in general trustworthy”: “strongly disagree” to “strongly agree”; M=0.69, SD=0.19); and commenters' age (young vs. older) were measured before any stimulus exposure and were included as covariates.

Analysis

First, the effects of others' opinion cues on both outcome variables were assessed using multiple regression analysis (Model 1). Next, the extent to which the effects were moderated by either IPSE or NFC (two-way interaction) was assessed using moderated ordinary least squares regression analysis (Model 2a and b). Finally, the extent to which the interaction effect of IPSE was moderated by NFC (three-way interaction) was analyzed (Model 3).

Results

Results of multiple regressions demonstrated that, as hypothesized, others' opinion cues were strong and significant predictors of perceived trustworthiness (H1a) and expertise (H1b) above and beyond the influences of participants' sex, race, frequency of Facebook use, ideology, political interest, general trust in politicians, and commenters' age (see Table 1). Thus, H1a and H1b were supported.

Table 1.

Hierarchical Regression Models Predicting Perceived Trustworthiness (N=251)

	Model 1		Model 2a (2-way)		Model 2b (2-way)		Model 3 (3-way)
	B	SE	B	SE	B	SE	B	SE
Controls
Sex	−0.05	0.17	−0.36	0.25	−0.35	0.25	−0.40	0.25
Race	0.15	0.18	−0.05	0.26	−0.10	0.27	−0.06	0.27
Ideology	−0.14	0.31	0.18	0.46	0.21	0.26	0.31	0.46
Pol interest	0.15	0.38	−0.22	0.58	−0.35	0.59	−0.41	0.60
Politician trust	−0.09^*	0.04	0.01	0.07	−0.01	0.07	−0.02	0.07
Facebook use	0.02	0.03	−0.02	0.05	−0.01	0.05	−0.003	0.05
Commenter age	−0.06	0.17	−0.24	0.25	−0.22	0.26	−0.17	0.26
Predictor
Others' opinions	3.00^***	0.17	4.59^***	0.93	3.58^**	1.15	11.61^***	3.42
2-way interactions
Others' opinions×IPSE			−2.56+	1.38			−13.72^*	5.29
Others' opinions×NFC					−1.14	1.96	−12.32^*	5.87
3-way interactions
Others' opinions×IPSE×NFC							19.30^*	8.78
Model R²	0.37^***		0.38^***		0.37^***		0.40^***
R² change			0.009+		0.001		0.02^*

***

p<0.001; ^**p<0.01; ^*p<0.05;+p<0.10.

IPSE, internal political self-efficacy; NFC, need for cognition.

Participants' IPSE moderated the effect of others' opinion cues on perceived trustworthiness, albeit marginally (H2a; see Table 1). The conditional effect of the cues was largest when IPSE was low (1 SD lower than the mean; b=3.39, SE=0.35, p<0.001), smaller when IPSE was at the mean (b=2.92, SE=0.25, p<0.001), and smallest when IPSE was high (1 SD higher than the mean; b=2.46, SE=0.36, p<0.001). IPSE significantly interacted with others' opinion cues in predicting perceived expertise (H2b; see Table 2). The conditional effect of others' opinion cues was largest when IPSE was low (b=2.68, SE=0.31, p<0.001), with the effect getting smaller when IPSE was at the mean (b=2.09, SE=0.22, p<0.001), and smallest when IPSE was high (b=1.49, SE=0.31, p<0.001). H3a and H3b hypothesized two-way interactions between NFC and others' opinions in predicting perceived trustworthiness and expertise. However, no evidence of such an interaction was found.

Table 2.

Hierarchical Regression Models Predicting Perceived Expertise (N=251)

	Model 1		Model 2a (2-way)		Model 2b (2-way)		Model 3 (3-way)
	B	SE	B	SE	B	SE	B	SE
Controls
Sex	0.03	0.15	−0.01	0.22	0.02	0.22	−0.05	0.22
Race	0.01	0.16	−0.40+	0.23	−0.51^*	0.23	−0.45	0.23
Ideology	0.04	0.28	−0.01	0.40	0.05	0.40	0.17	0.39
Pol interest	0.59+	0.34	0.05	0.51	−0.21	0.51	−0.28	0.51
Politician trust	−0.02	0.04	0.07	0.06	0.04	0.06	0.03	0.06
Facebook use	0.04	0.03	0.02	0.04	0.03	0.04	0.04	0.04
Commenter age	−0.07	0.15	−0.21	0.22			−0.11	0.22
Predictor
Others' opinions	1.99^***	0.15	4.22^***	0.81	2.59^*	1.0	11.97^***	2.93
2-way interactions
Others' opinions×IPSE			−3.29^**	1.21			−16.12^***	4.5
Others' opinions×NFC					−0.86	1.71	−13.48^**	5.02
3-way interaction
Others' opinions×IPSE×NFC							22.5^**	7.51
Model R²	0.29^***		0.31^***		0.31^***		0.35^**
R² change			0.02^**		0.001		0.04^**

***

p<9.001; ^**p<0.01; ^*p<0.05;+p<0.10.

As for RQ1a, a significant three-way interaction among others' opinion cues, IPSE, and NFC was found in predicting perceived trustworthiness (see Table 1). The conditional two-way interaction effect between others' opinion cues and IPSE was strongly significant when NFC was low (b=−5.20, SE=1.87, p=0.006), marginally significant when NFC was at the mean (b=−2.69, SE=1.41, p=0.058), and nonsignificant when NFC was high (b=−0.18, SE=1.77, p=0.920). Others' opinion cues had the largest influence on perceived trustworthiness in participants with low IPSE and low NFC (b=3.74, SE=0.43, p<0.001). The smallest effect was found when IPSE was high and NFC was low (b=1.85, SE=0.56, p=0.001), not when both were high (b=2.82, SE=0.44, p<0.001). When IPSE was at the mean, the conditional effect of comment valence was larger when NFC was high (b=2.85, SE=0.36, p<0.001) than when NFC was low (b=2.80, SE=0.37, p<0.001).

A significant three-way interaction among others' opinion cues, IPSE, and NFC was found in predicting perceived expertise, thereby supporting RQ1b (see Table 2). The two-way interaction effect between others' opinion cues and IPSE was strongest when NFC was low (b=−6.39, SE=1.60, p<0.001), significant but smaller when NFC was at the mean (b=−3.52, SE=1.21, p=0.004), and nonsignificant when NFC was high (b=−0.64, SE=1.51, p=0.671). Others' opinion cues had the largest influence on perceived expertise in participants with low IPSE and low NFC (b=3.03, SE=0.31, p<0.001); the effect was nonsignificant (i.e., smallest) when IPSE was high and NFC was low (b=0.71, SE=0.48, p=0.142). When IPSE was at the mean, again, the conditional effect of others' opinion cues was larger when NFC was high (b=2.09, SE=0.31, p<0.001) than when NFC was low (b=1.87, SE=0.32, p<0.001). These results were found with participants' sex, race, frequency of Facebook use, ideology, political interest, trust in politicians, and commenters' age controlled. In the final three-way interaction models, no control variable was significant as a predictor of either trustworthiness or expertise.

Discussion

As SNS have become increasingly important sources of political information, it is crucial to investigate the mechanisms by which individuals use the information. In particular, it is important to address how individuals differently evaluate an unknown person based on available cues. The present study investigated the role of IPSE and NFC relative to others' opinion cues in impression formation of a political candidate's trustworthiness and expertise. As expected, when there was no other information or preformed attitude toward a person, others' comments were powerful cues. This result might help explain why politicians who gained success on SNS were able to succeed in elections. However, the extent to which participants relied on the cues differed depending on their IPSE and NFC.

Overall, the higher people's IPSE was, the less susceptible they were to others' opinions. It means that IPSE influenced the extent to which they depended on the cues. This result adds nuance to Holbert et al.'s finding that IPSE can function as a key motivational factor that moderates the effect of political messages. However, NFC did not significantly moderate the cue effect. Participants similarly relied on the comment cues in gauging a candidate's trustworthiness and expertise regardless of their NFC. This result might be related to the design of the experiment. High-NFC individuals tend to base their judgments on the argument quality of persuasive messages whereas low-NFC individuals heavily rely on heuristic cues.²⁹ In the current experiment, others' comments were the most relevant information about the credibility of the candidate within the environment, and could also function as substantial arguments for high-NFC people, while they functioned as a heuristic cue for low-NFC people, thereby mitigating the differences between high- and low-NFC people.

Finally, NFC's moderating role was significant in the test of three-way interaction; the two-way interaction effect between IPSE and others' opinion cues significantly varied by the level of NFC. Others' comments exerted the greatest effect over those whose IPSE and NFC were low, whether it was about the perception of trustworthiness or expertise. This result indicates that those who dislike cognitive activities and lack confidence in understanding and making an impact on politics are most likely to rely on available others' comments as cues. It is notable, however, that among people with high IPSE, the lower NFC they had, the less they were susceptible to others' influences. It is counterintuitive that those with high IPSE and low NFC, not high NFC, were most resistant to others' influences. It might be the case that while high-IPSE people would avoid relying on others' opinion cues too much, if they have high NFC, they would take the comments as a substantial message upon which to base their judgment. On the other hand, low-NFC people might just follow the guidance of IPSE (i.e., avoid making a decision simply based on others' comments) rather than actively trying to solve a problem in the given situation. Further investigation is needed to shed light on the underlying mechanism.

Several limitations of this study should be acknowledged. First, the sample consisted entirely of young college students, thereby limiting the applicability of the results beyond this age group. Nevertheless, this sample served well for addressing how young people with varying levels of IPSE and NFC would react to others' opinion cues, which is the topic of interest. Second, no information about the candidate's background, issue positions, or even party affiliation was provided in the stimuli, whereas voters can access a lot more information in the real world. We intentionally removed those elements from the stimulus to focus on the effect of others' opinions while minimizing potential confounders. This setting is artificial but not seriously unrealistic given that impressions are always formed based on a few immediately available cues without careful contemplation of all relevant factors. Moreover, comprehensive summary information or even party affiliation is not always available on politicians' SNS pages. Thus, the findings of this study are still widely applicable. Nonetheless, future studies will benefit from building on this study and replicating it in the more complex environment where multiple cues compete with one another.

The present study contributes to the field as one of the few attempts to examine the roles of individual psychological and motivational factors in regards to others' influences in impression formation. This study provides a bridge between CMC and political communication by testing individual differences in the context of a political judgment on a political candidate's social media page. It is a meaningful bridge because CMC traditionally has its roots in interpersonal communication, whereas political communication has been studied mainly within the context of mass communication. The findings add details to the SIP framework; not only does the situation of communication (e.g., face-to-face vs. computer mediated) matter, but so do individual psychological and motivational factors. The use of experimental design and multiple stimuli enabled the testing of causal relationships and provided better generalizability of the results.

This study also has practical implications. Young adults are typically known to lack the motivation and ability to engage in politics and public affairs³² The finding that those with low IPSE and NFC are easily swayed by others' opinion cues on SNS indicates how serious the situation can become when there are attempts to manipulate “others' opinion cues” on social media sites. It is important for SNS users to realize how easily their perceptions and evaluations of politicians can be influenced by several unrepresentative comments of others with dramatic consequences for our politics. These findings can be used to boost young voters' levels of political motivation. While many political communicators treat SNS users as a group that is uniformly influenced by their messages, communication will be more effective when they consider the general interest or motivation levels of the target population and approach accordingly. Scholars are encouraged to examine further other individual differences that may make people more susceptible or resistant to others' influences on SNS.

Footnotes

Acknowledgments

The author would like to thank Young-shin Lim at The Ohio State University for her contribution to the data collection, and the three anonymous reviewers, particularly Reviewer #3, for their insightful comments and suggestions.

Author Disclosure Statement

No competing financial interests exist.

References

Howard

, Parks

. Social media and political change: capacity, constraint, and consequence. Journal of Communication, 2012; 62:359–362.

Kennedy

. (2013). Use it or lose it: social media in the 2012 U.S. Election. http://pulitzercenter.org/reporting/social-media-role-young-voters-increase-future-US-elections-Obama-Facebook-Twitter (accessed July 1, 2014).

Williams

, Gulati

. (2007). What is a social network worth? Facebook and vote share in the 2008 presidential primaries. Paper presented at the annual conference of the American Political Science Association, Boston, MA.

Lee

, Oh

. To personalize or depersonalize? When and how politicians' personalized Tweets affect the public's reactions. Journal of Communication, 2012; 62:932–949.

Walther

, Van Der Heide

, Kim

, et al. The role of friends' behavior on evaluations of individuals' Facebook profiles: are we known by the company we keep?. Human Communication Research, 2008; 34:28–49.

Walther

, Van Der Heide

, Hamel

, et al. Self-generated versus other-generated statements and impressions in computer-mediated communication: a test of warranting theory using Facebook. Communication Research, 2009; 36:229–253.

Chaiken

, Liberman

, Eagly

. (1989) Heuristic and systematic information processing within and beyond the persuasion context. In Uleman

, Bargh

, eds. Unintended thought. New York: Guilford Press, pp. 212–252.

Simon

. (1976) Administrative behavior. 3rd ed. New York: Free Press.

Chen

, Chaiken

. (1999) The heuristic-systematic model in its broader context. Dual-process theories in social psychology. New York: Guilford Press, pp. 73–96.

10.

Tversky

, Kahneman

. Judgment under uncertainty: heuristic and biases. Science, 1974; 185:1124–1131.

11.

Walther

. Interpersonal effects in computer-mediated interaction: a relational perspective. Communication Research, 1992; 19:52–90.

12.

Brunswik

. (1956) Perception and the representative design of psychological experiments. Berkeley: University of California Press.

13.

Lee

, Jang

Y-J

. What do others' reactions to news on Internet portal sites tell us? Effects of presentation format and readers' need for cognition on reality perception. Communication Research, 2010; 37:825–846.

14.

Sundar

, Nass

. Conceptualizing sources in online news. Journal of Communication, 2001; 51:52–72.

15.

Hovland

, Janis

, Kelley

. (1953) Communication and persuasion: psychological studies of opinion change. New Haven, CT: Yale University Press.

16.

McCroskey

. Scales for the measurement of ethos. Speech Monographs, 1966; 33:65–72.

17.

Ohanian

. Construction and validation of a scale to measure celebrity endorser's perceived expertise, trustworthiness and attractiveness. Journal of Advertising, 1990; 19:39–52.

18.

Perloff

. (2003) The Dynamics of persuasion: communication and attitudes in the 21st century. 2nd ed. Mahwah, NJ: Lawrence Erlbaum.

19.

Frensch

, Sternberg

. (1989) Expertise and intelligent thinking: when is it worse to know better? In Sternberg

, ed. Advances in the psychology of human intelligence (Vol. 5.). Hillsdale, NJ: Lawrence Erlbaum, pp. 157–188.

20.

Crano

. Effects of sex, response order, and expertise in conformity: a dispositional approach. Sociometry, 1970; 33:239–252.

21.

Crisci

, Kassinove

. Effect of perceived expertise, strength of advice, and environmental setting on parental compliance. The Journal of Social Psychology, 1973; 89:245–250.

22.

Slepian

, Young

, Rule

, et al. Embodied impression formation: social judgments and motor cues to approach and avoidance. Social Cognition, 2012; 30:232–240.

23.

Kunda

. The case for motivated reasoning. Psychological Bulletin, 1990; 108:480–498.

24.

Niemi

, Craig

, Mattei

. Measuring internal political efficacy in the 1988 National Election Study. The American Political Science Review, 1991; 85:1407–1413.

25.

Craig

. Efficacy, trust, and political behavior: an attempt to resolve a lingering conceptual dilemma. American Politics Quarterly, 1979; 7:225–239.

26.

Vecchione

, Caprara

. Personality determinants of political participation: the contribution of traits and self-efficacy beliefs. Personality & Individual Differences, 2009; 46:487–492.

27.

Holbert

, Lamb

, Dudo

, et al. Primacy effects of the Daily Show and national TV news viewing: young viewers, political gratifications, and internal political self-efficacy. Journal of Broadcasting & Electronic Media, 2007; 51:20–38.

28.

Cohen

, Stotland

, Wolfe

. An experimental investigation of need for cognition. The Journal of Abnormal & Social Psychology, 1955; 51:291–294.

29.

Cacioppo

, Petty

, Feinstein

, et al. Dispositional differences in cognitive motivation: the life and times of individuals varying in need for cognition. Psychological Bulletin, 1996; 119:197–253.

30.

Petty

, Cacioppo

. (1986) Communication and persuasion: central and peripheral routes to attitude change. New York: Springer.

31.

Arceneaux

, Vander Wielen

. The effects of need for cognition and need for affect on partisan evaluations. Political Psychology, 2013; 34:23–42.

32.

Delli Carpini

. Gen.com: youth, civic engagement, and the new information environment. Political Communication, 2000; 17:341–349.