Changing attitudes and intentions related to doping: An analysis of individual differences in need for cognition

ABSTRACT

Research has shown that athletes’ attitudes towards the use of banned performance-enhancing substances are reliable predictors of their intentions to use these substances, which in turn can be relevant predictors of their actual doping behaviours. Despite the important role played by attitudes and intentions in doping, research analysing how to change those attitudes and intentions is relatively scarce. The present study examined how individual differences in Need for Cognition (NC, Cacioppo & Petty, 1982) influenced doping-related attitude change and subsequent behavioural intentions. Participants were randomly assigned to read a persuasive message either against or in favour of legalising the use of several banned substances, including anabolic androgenic steroids (AAS) and Erythropoietin (EPO). In addition, participants completed the NC scale, and finally reported their attitudes and behavioural intentions regarding the legalisation proposal. As hypothesised, results showed that participants who received an anti-legalisation message had significantly more unfavourable attitudes towards the proposal than participants who received a pro-legalisation message, regardless of NC. However, as predicted, NC moderated the relationship between individuals’ attitudes and their intentions. That is, high-NC participants showed greater attitude-intention correspondence than low-NC participants.

KEYWORDS:

• Need for cognition
• attitude
• behavioural intention
• doping
• elaboration

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ It is important to note that low-NC individuals can be motivated to elaborate on the message content (although they are not naturally inclined to do so) if other factors are present, such as when an issue is personally relevant, or when they are held responsible for subsequent outcomes. Likewise, high-NC individuals may be unable to elaborate on the message content and thus base their attitudes on peripheral cues, such as when insufficient time is available to process a message.

² Participants were required to meet the following inclusion criteria: a) fall between 18 and 26 years of age; b) engage in sports or physical exercise at least 3 hours per week; and c) not be enrolled as last-year university students because these students could have already taken upper-year attitudes and persuasion courses and thus be aware of our hypotheses, which could potentially bias their responses.

³ Participants were told that messages did not include rigorous and accurate scientific information (most importantly, about the certain benefits of those substances and behaviours). Furthermore, a message in favour of legalisation was included to have more equivalent tasks across experimental conditions and to permit a better comparison with a message against legalisation (Horcajo & De la Vega, 2014). Applications of this research (e.g., doping prevention programmes) should include the implementation of effective programmes to change doping-related attitudes in only one direction (i.e., against doping).

⁴ These variables are expected to serve as simple cues when recipients are under low-elaboration conditions (such as low-NC individuals). However, these same variables can also affect attitudes through different processes under other circumstances (e.g., see Petty & Briñol, 2012; Petty & Cacioppo, 1986; Petty & Wegener, 1998).

⁵ To control for the potential effect of behavioural intentions on attitudes, participants’ intentions were included in the analysis as a covariable.

⁶ When Message Direction and NC were the only predictors included in the model, the main effect of Message Direction remained significant, B = 1.650 (95%CI: 1.242, 2.057), t(133) = 8.014, p < .001, Cohen’s f = 0.695, indicating that attitudes were more unfavourable for the anti- than pro-legalisation message. No other effects were significant (ps > .12).

⁷ When Message Direction and NC were the only predictors included in the model, a main effect of Message Direction emerged, B = .946 (95%CI: .517, 1.375), t(133) = 4.358, p < .001, Cohen’s f = 0.378, indicating lower behavioural intentions for the anti-legalisation than pro-legalisation message. Also, a main effect of NC emerged, B = .361 (95%CI: .001, .721), t(132) = 1.985, p = .049, Cohen’s f = 0.172, indicating greater behavioural intentions as NC increased.

⁸ A marginally significant three-way interaction (Message Direction × NC × Attitudes) emerged, B = .515 (95%CI: −.029, 1.078), t(132) = 1.877, p = .063, Cohen’s f = 0.166, indicating that the expected pattern of results was more pronounced for the pro-legalisation (vs. anti-legalisation) message. In addition, there was a marginal effect of Message Direction × NC interaction, B = .639 (95%CI: −.081, 1.359), t(132) = 1.756, p = .081, Cohen’s f = 0.153. In line with prior research (Horcajo & Luttrell, 2016), for the high-NC participants, those who received the anti-legalisation message reported significantly lower intentions to support legalisation and to engage in doping behaviours if legalised than those who received the pro-legalisation message, B = 1.326 [95%CI: .722, 1.929], t(132) = 4.344, p < .001, Cohen’s f = 0.379. For the low-NC participants, there was only a marginal difference between those who received the message against and those who received the message in favour of legalisation, B = .563 [95%CI: −.044, 1.169], t(132) = 1.836, p = .069, Cohen’s f = 0.160. Finally, there was no Message Direction × Attitudes interaction, B = .056 (95%CI: −.256, .369), t(132) = .356, p = .722, Cohen’s f = 0.031.

⁹ According to the ELM (e.g., see Petty & Briñol, 2012), when either motivation or ability to think is low, variables typically serve as simple peripheral cues, producing an effect that is consistent with their valence (i.e., variables evaluated positively lead to more persuasion). When motivation and ability to think are high, variables can produce persuasion via other processes, such as serving as arguments (pieces of evidence), biasing thoughts, or influencing the degree to which a person has confidence in their thoughts (i.e., self-validation) and thus relies on them when making judgements. When thinking is not constrained to be high or low by other factors, then variables determine how much thinking is done.

Additional information

Funding

This research was supported by the Ministerio de Economía, Industria y Competitividad, Gobierno de España (ES) [PSI2017-83303-C2-1-P] Grant to Javier Horcajo.