Dissertation on Demographic Credibility in Communication

Info: 10059 words (40 pages) Dissertation
Published: 17th Nov 2021

Reference this

Tagged: CommunicationsEquality

Share this: Facebook Twitter Reddit LinkedIn WhatsApp

Click to expand Table of Contents

Introduction

Communication of science

Environmental communication

Credibility of the information source

Occupation

Gender

The present study

Hypotheses

Method

Participants

Apparatus

Materials

Design

Procedure

Results

Attention checks

Credibility rating

Meat eating behavioural intention

Energy provider behavioural intention

Aeroplane travel behavioural intention

Discussion

Practical limitations

Implications for policy and practice

Future research

Conclusion

References

Appendices

Appendix 1

Appendix 2

Appendix 3

Introduction

Communication of science

Communication is central to knowledge and understanding of scientific issues. Scientists have the responsibility to create and contribute ideas to a consensus of public information (Ziman et al., 1969). For society to be democratic, experts must present specialised knowledge into public domain so that non-experts can make better-informed opinions and decisions (cited in Improving Risk Communication, 1989). The way in which information is communicated in the public sphere is important because the interpretation of the message can influence perception and subsequently behaviour (Hansen, 2011). It can be challenging to facilitate public understanding in science communication, as it is often accompanied by public distrust. In particular, evidence suggests a long term decline of public confidence in typical sources such as the government and industry (Peters et al., 1997; Slovic 1993). As stated by the National Research Council (1989), effective communication is important to minimise the public distrust and misunderstanding that might occur in response to poor communication.

Environmental communication

Despite an abundance of scientific evidence, mass media coverage and environmental policies, many people remain sceptical of certain scientific topics deemed controversial, including climate change (Slovic, 1993; Whitmarsh, 2011). Even with 97% of environmental scientists agreeing that human activity is contributing to climate change (Doran and Zimmerman, 2009), public beliefs of climate change have continued to be uncertain (Renn and Levine, 1991; Hulme, 2009). Climate change cannot be directly experienced therefore it is necessary to trust the sources who impart the information (Spengler, 1932).

Even with scientists communicating that global warming is perhaps the most significant environmental issue of our time, it remains low on the public list of concern (Poortinga and Pidgeon, 2003; Konisky et al., 2008; Sandvick, 2007), with only one third of the public taking action ‘out of concern for climate change’ (Whitmarsh, 2009). Much of climate change information is conveyed to the public through mass media (Aoyagi-Usui, 2008; Eurobarometer, 2001), thus trust and credibility of sources are significant mediators of public perception of climate change (Whitmarsh, 2005). Public perception is critically important to subsequently effect consumer behaviour and influences on policy. Hansen (2011) contends that “battles over these issues are now as much to do with communication aimed at winning hearts and minds as they are to do with communicating science based or expert evidence” (p.8). Evidence suggests that even when media coverage of climate change was at an all-time high, the level of public concern about climate change remained low (Leiserowitz et al., 2013).

For example, Konisky et al. (2008) showed that participants perceived global warming as the eighth most important of twelve environmental issues. This indicates a considerable level of public distrust about the sincerity of climate change, and demonstrates how communicating environmental issues has the potential to divide public attitudes. This highlights the gap between public acceptability and climate change communications, and emphasises the importance of having trustworthy sources to communicate effective messages to the public.

Furthermore, Konisky et al. (2008) found that participants who had higher levels of trust in the source (government) were more likely to support government action in regards to addressing climate change. This implies that trust in the source is key for environmental communication to facilitate credulous attitudes and opinions of the public, which subsequently have a significant influence on behavioural intentions. With global warming perceived as being of low importance, the results of Konisky et al. (2008) suggest that having trust in the source is a significant predictor of public acceptability and engagement in environmental communication. Consequently, finding ways to maintain trust in sources should be a priority of the scientific community (Leiserowitz et al. 2013), in order for the crux of public engagement to occur before it’s too late (Leiserowitz, 2005).

Credibility of the information source

A key component of environmental communication is trust and credibility of the source (Slovic, 1993; Johnson, 1999; Earle and Cvetkovich, 1995). Institutions involved in communicating scientific issues in the public sphere often rely on trust and credibility as a powerful social resource in their communication technique (Renn and Levine, 1991), which implies the importance of source credibility in environmental communication. Given the importance of these factors in mediating environmental attitudes and behaviours, it is essential to know which elements facilitate trust and credibility. Peters et al. (1997) investigated the composite of credible sources and found it was based upon three factors described as; “knowledge and expertise, openness and honesty, and concern and care”. These factors reflect the extent to which the source is capable, honest and acting in accordance of public interest (Johnson, 1999). The determinants of credibility subsequently affects the level of trust in scientific communication, the perception of risk and thus behavioural intentions (Whitmarsh, 2005).

Occupation

Previous literature have inferred that the public tend to trust some sources more so than others. Some evidence suggests long term decline of trust in traditional sources such as the government and industry (Lipset and Schneider, 1983). In addition, believability of journalists in the press has dramatically declined over recent decades (Pew, 2005). Poortinga et al. (2004) investigated public risk perceptions and trust in government policy in regards to the 2001 foot and mouth crisis; when asked to state the extent to which they trusted various sources to inform them about the outbreak, participants had the most trust in veterinary surgeons and farmers. In contrast, participants rated government ministers and the European Union as the least credible.

Similarly, Leiserowitz et al. (2013) asked participants the extent to which they trusted or distrusted each of nine sources of information about global warming, with results showing that participants trusted scientists much more than mainstream media, weather reporters and even President Obama. Further evidence supports the idea that scientists are generally more trusted than other sources; in their survey, Poortinga and Pidgeon (2003) investigated public trust in different sources of climate change communication. Participants were asked to rate the extent to which they trusted each source to tell them the truth about climate change. Scientists working for universities and environmental organisations were among the most credible sources, in comparison to the national government and the European Union whom were regarded as least credible. In consideration of previous literature, the present study will include a scientist and a journalist from mass media as sources. The term source being used in reference to the person who wrote the article.

Gender

Earlier research has focussed on the credibility of sources in relation to gender. Evidence from previous research on gender and source credibility suggests that male sources are perceived as more credible than female counterparts. Pearson (1982) conducted telephone interviews consisting of four closed questions in which participants were asked overtly about gender differences in credibility, for example; ‘if you were the opposite sex, would you be viewed as having a) more credibility or, b) less credibility?’

Overall, men were perceived as more credible; although it could be argued that the study involved leading questions and therefore the results were limited. Results from previous research have illustrated that people use the perceived credibility of the source as a heuristic to decide whether to accept the message (Cacioppo et al., 1986; Petty et al. 1981), in other words, schemas and stereotypes are relied upon to inform judgments.

In recent literature, Armstrong and McAdams (2009) demonstrated how gender cues might activate stereotypes and influence perceived credibility of informational blogs. The participants, whom were undergraduate students, read a blog post of either a male, female or unspecified author and asked to rate the credibility of the source. Blogs perceived to be written by male authors were rated higher in credibility. This suggests that gender is used as a heuristic to determine credibility.

Research by Armstrong and Nelson (2005) explains why gender stereotypes in relation to credibility of an information source might occur. They discuss that official sources in the media have most often been male (Ross and Carter, 2011), and as the media plays a significant role in gender socialisation, with men appearing more frequently as information sources than women, this might lead to the formation of gender stereotypes (Armstrong and Nelson, 2005).

In Armstrong and Nelson’s (2005) study, when shown an expert source where gender was made ambiguous, participants tapped into their stereotyping heuristics, assuming the source was male. Gender stereotypes may be acquired by seeing the representation of men and women in certain occupations. Evidence of gender stereotypes have been shown in children; when asked to draw a scientist, only five of thirty participants drew a female (Buldu, 2006). This may be a reflection of the imbalance of gender that currently exists in some particular occupations; for example just 24% of the STEM workforce in the UK is female (WISE, 2017), the imbalance is also reflected subject choice. Science subjects persist to be dominated by male students in schools and higher education, where only 9% of female students go on to study STEM degrees compared to 29% of male students (WISE, 2017). This implies that gender norms have been perpetuated through societal generations.

Further evidence of inequality in society has been found in previous research; Steinpreis et al. (1999) found that male applicants were more likely to be hired by academics despite an identical CV used to represent female applicants.

More recently, evidence of inequality that has been covered in current media is the gender pay gap, which may be a result of socialisation of gender norms. The gender pay gap is the difference in average pay between men and women. The Equality Act 2010 (Gender Pay Gap Information) Regulations 2017 requires businesses with more than 250 employees to report gender pay gap statistics, which are made publicly available on the government website.

Figures shown on the gender pay gap service (GOV.UK, 2017) demonstrate that men dominate the top quartile of pay in almost all professions, including science. A gender pay gap is present even in UK universities including Cardiff University; figures show that on average, women earn 21.8% less per hour than men (mean gender pay gap) (Cardiff University, 2017).

Current gender inequalities have been highlighted in recent research and therefore prompts the question if men would be regarded as more credible than female counterparts if they were the source of scientific information.

The present study

The present study aims to look at the importance of the source in climate change communication. In particular, how certain characteristics of an information source might prompt environmentally friendly attitudes and behavioural intentions.

The present study aims to investigate how the differences between occupation and gender of authors might influence participants’ willingness to engage in environmentally friendly behaviour, to implicate what constitutes effective environmental communication.

This will be studied by measuring and comparing the differences between behavioural intentions of participants, whom will be presented with information from one of four sources, each differing in occupation and gender.

Hypotheses

In consideration of the results from previous research, it is predicted that scientists will be perceived as significantly more credible than journalists and will elicit greater willingness to participate in environmentally-friendly behaviour. Secondly, it is predicted that male authors will be regarded as significantly more credible than their female counterparts and as a result, participants in experimental conditions with a male author will be more willing to engage in environmentally friendly behaviour. Finally, it is hypothesised that male scientists will be perceived as the most credible author and female journalists seen as the least credible.

Method

Participants

The participants (n = 119) were recruited via the Cardiff University School of Psychology Experimental Management System. All participants were aged between 18 and 24. The sample population consisted of first and second year Psychology undergraduate students, who gained course credit in exchange for their participation. Of the whole sample, 105 (88.2%) of the participants were females and 13 (10.9%) of the participants were males, and 1 (0.8%) preferred not to state their gender.

The participants were randomly assigned by Qualtrics software (Qualtrics, Provo, UT) into one of four experimental groups. One participant failed to attempt completion of the study therefore data of 118 responses were recorded; thus the Male Scientist condition and the Male Journalist condition each had 29 assigned participants. The Female Scientist condition and the Female Journalist condition each had 30 participants.

Apparatus

The study was accessible to participants via an online URL presented on the School of Psychology Experimental Management System. The participants completed the study on their own personal devices before a set deadline. Qualtrics software ensured the participants could not go backwards to a previous part of the experiment, for example so that they could not read the article again. The data was analysed using SPSS software.

Materials

The experiment involved a fictitious online article that was written in the perspective of either an environmental journalist from a newspaper or an environmental scientist from a university (see Appendix 1). The author suggested three pro-environmental behavioural adjustments which were specifically selected on the basis that they would be perceived as challenging. This was to avoid a ceiling effect had behaviours been selected that were deemed as too easy.

All components of the article were fictional and created on Microsoft Word. The names presented, Simon/Rosie Jenkins, were not genuine authors. In addition, the places of work that were stated; University of Lancaster/Daily Review, were also imaginary. Pictures of the supposed authors were selected from the ‘free to use or share’ section on Google Images. Images were sourced based upon similarity of age, in an attempt to control for any confounding variables.

A slide was created with the purpose of making the gender and place of work salient before the participants were presented with the article, it would read for example “On the next page, you will see an article written by Dr. Simon Jenkins from the University of Lancaster.” The article and its associated source were used to investigate the differences between perceived credibility of the author and how this might alter behavioural intentions.

Attention checks were created to ensure that participants had processed the information of the article, and to foster internal validity. The attention checks consisted of 8 multiple choice questions which tested comprehension of the information presented in the article (see Appendix 2). There were 2 attention checks that were of utmost importance, in which the participants were asked “What was the gender of the author?”, and “What was the occupation of the author?”

The data from participants who answered these attention checks incorrectly would be removed since they would not have processed the information about the author and as a result, any effect on the dependent variables might not be valid for the present study.

Included in the study was a list of twenty-five statements to measure behavioural intentions. All statements were written in a pro-environmental direction. Each statement was accompanied by a seven-point Likert scale ranging from (1) ‘strongly disagree to (7) ‘strongly agree’. The participants were asked about the extent to which they agreed with the statements, which were written in first person and measured willingness to engage in certain behaviours, for example;

“I will reduce my weekly meat consumption.”
“I will research several methods of transport before my next holiday.”
“I will talk to others about which energy provider they use.”

Out of all the statements, responses to only twelve questions were of interest (see Appendix 3)because these statements focussed on the three behaviours that were suggested by the author of the article. Four statements measured the intention of each specific behaviour. Reliable scales were formed to measure the behavioural intentions of the three particular behaviours; meat eating behavioural intention (Cronbach’s Alpha = .88), energy provider behavioural intention (Cronbach’s Alpha = .90) and aeroplane travel behavioural intention (Cronbach’s Alpha = .78).

Design

The experiment used a between-subjects design. Two independent variables were manipulated between-subjects. The first independent variable was the gender of the author (male or female). The second independent variable was the occupation of the author, which also had two levels; newspaper environmental journalist from the ‘Daily Review’ and university environmental scientist from the ‘University of Lancaster’. There were four experimental conditions: Male Scientist, Female Scientist, Male Journalist and Female Journalist.

The Qualtrics software randomly assigned participants to one of four conditions. The content of the article remained the same throughout all four conditions, only the picture, name, occupation and place of work was manipulated in the article.

The experiment had four dependent variables, which were the credibility of the author, meat eating behavioural intention, energy provider behavioural intention and aeroplane travel behavioural intention. The participants rated the credibility of the author on a five-point Likert scale ranging from (1) ‘not at all credible’ to (5) ‘extremely credible’. All three behavioural intention dependent variables were measured by seven point Likert scales ranging from (1) ‘strongly disagree’ to (7) ‘strongly agree’.

The Qualtrics software enabled answers to the attention checks to be presented in a randomised order. The behavioural intention items were also presented in a random order.

Procedure

The participants accessed the online study through the School of Psychology Experimental Management System. A consent form was presented before the experiment which informed the participants what they would be expected to do without disclosing the aim of the study. All participants were informed of their right to withdraw and the anonymity of the data. The study would not run unless participants had selected the consent button. An estimate of fifteen minutes was given for the completion of the study.

Next, a short demographics survey was used to collect data in regards to age and gender. The participants were then shown a slide which contained information about whom had written the article they were about to be presented with. All participants were shown an online article written by one of four authors, depending on which condition the participant had been assigned to.

Following this, the participants were asked to complete eight multiple choice attention check questions to assess their comprehension of the article. Lastly, twenty-five behavioural intention statements were presented, all of which were related to environmental behaviour. Participants were asked to state the extent to which they agreed with each statement by selecting one of seven points on a Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’.

A debrief form was shown to participants which explained the aims of the study and fictitious nature of the materials. The participants were given the contact details of the researcher and supervisor should they have any concerns or questions about the study. Upon completion, participants gained partial course credit.

Results

SPSS software was used to analyse the raw data. Two-way between subjects ANOVAs were performed to compare the differences between source characteristics, notably gender and occupation, and behavioural intentions. Behavioural intentions were measured for three behaviours consisting of: meat eating behavioural intention, energy provider behavioural intention and aeroplane travel behavioural intention.

Attention checks

For analysis purposes, participants who gave incorrect answers to the gender and occupation attention checks were removed from the data set as it was imperative for these to be answered correctly in the current study. 2.5% (n = 3) of participants answered the gender attention check incorrectly.11.8% (n = 14) of participants gave an incorrect answer to the occupation attention check. The analysis was run after said participants had been removed from the data set however no significant main effects or interactions were subsequently found. Consequently, the original data set that included responses from all participants was used in the analysis of the present study.

Credibility rating

The mean and the standard deviation of rating of credibility was recorded for each of the experimental conditions, and are displayed in the figure below (see Figure 1). As Figure 1 shows, the mean of rated credibility was highest in the Female Scientist condition: 3.53.The lowest number of mean credibility rating was recorded in the Male Scientist condition. An inferential statistical test was used to assess the significance of the results.

Figure 1. Mean credibility rating of the author in each experimental condition.

In order to assess the differences between source characteristics and credibility rating, a two-way between subjects ANOVA was performed. The results of the experiment show that the main effect of source gender on perceived credibility was non-significant (F(1, 111) = .13, p > .05, MSE = .48). The results also show a non-significant main effect of source occupation on perceived credibility (F(1, 111) = .001, p > .05, MSE = 1). The interaction between gender and occupation was also found to be non-significant (F(1, 111) = .98, p > .05, MSE = .48), thus the results indicate that there is no relationship between the gender or occupation of the source and perceived credibility.

Meat eating behavioural intention

The means and standard deviations of intention to engage in pro-environmental meat eating behaviour were recorded for each condition (see Figure 2). Figure 2 presents the means and standard deviations for the four conditions. The mean score of behavioural intention was highest in the Male Scientist condition: 4.63. In addition, the group that had the lowest mean behavioural intention score was the Female Journalist condition. An inferential statistical test was carried out to examine the significance of these results.

Figure 2. Mean meat eating behavioural intention score of each experimental condition.

The goal of assessing the differences between meat eating behavioural intentions of each condition was realised by performing a two-way between subjects ANOVA. The results show that the main effect of gender on meat eating behavioural intention was non-significant (F(1, 112) = .77, p > .05, MSE = 2.6). Similarly; the results illustrate that the main effect of source occupation on meat eating behavioural intention was found to be non-significant (F(1, 112) = .001, p > .05, MSE = 1). The interaction between gender and occupation of the source was also shown to be non-significant (F(1, 112) = .98, p > .05, MSE = .48). It could therefore be suggested that there is no relationship between source characteristics and behavioural intentions in regards to eating meat.

Energy provider behavioural intention

The means and standard deviations of behavioural intention in relation to pro-environmental energy providers were recorded for all four conditions, and are presented in the figure below (see Figure 3). As seen in Figure 3, the mean score for behavioural intention was greatest in the Male Journalist condition (4.63). Moreover, behavioural intention was least in the Female Journalist condition.

Figure 3. Mean energy provider behavioural intention score of each experimental condition.

A two-way between subjects ANOVA was completed to assess the differences between source characteristics and behavioural intention in regards to energy providers. The results show that a main effect of gender on energy provider behavioural intentions was non-significant (F(1, 112) = 1.46, p > .05, MSE = 1.85). Subsequently, the results show a non-significant main effect of occupation on behavioural intention (F(1, 112) = .23, p > .05, MSE = 1.85). The interaction between gender and occupation of the source was also found to be non-significant (F(1, 112) = .82, p > .05, MSE = 1.85), hence the results indicate that there is no relationship between source characteristics and behavioural intentions related to energy providers.

Aeroplane travel behavioural intention

The means and standard deviations of intention to adopt environmentally friendlier aeroplane travel behaviour are presented in Figure 4. As seen in Figure 4, the mean score of intention to engage in environmentally friendly travel behaviour was highest in the Male Journalist condition: 4.14 and lowest in the Female Scientist condition.

Figure 4. Mean aeroplane travel behavioural intention score of each experimental condition.

A two-way between subjects ANOVA was calculated to evaluate the differences between source characteristics and behavioural intentions in respect to aeroplane travel. The results display a non-significant main effect of gender on aeroplane travel behavioural intentions (F(1, 112) = .27, p > .05, MSE = 1.44). In addition, the findings show a non-significant main effect of occupation on aeroplane travel behavioural intentions (F(1, 112) = .25, p > .05, MSE = 1.41). The interaction between gender and occupation was shown to be non-significant (F(1, 112) = .05, p > .05, MSE = 1.43), therefore the results suggest that there is no relationship between source characteristics and behavioural intentions related to aeroplane travel.

It can therefore be concluded that there was no significant difference between each of the experimental conditions. In particular, neither one of four sources in possession of particular characteristics was perceived to be more credible than its counterparts.

Discussion

The results showed that no significant differences were realised between gender of the author and behavioural intentions, in regard to the three suggested behaviour adaptions.

Similarly, no significant differences were found between the occupation of the author and behavioural intentions. Interactions between gender and occupation variables, in relation to each behavioural intention measure, were shown to be non-significant.

The Male Scientist was not perceived as significantly more credible than any of the other authors. In addition, the Female Journalist was not perceived as significantly less credible than the other sources.

The results indicate that scientists were not perceived as more credible than journalists, nor males as more credible than females, which permits all hypotheses to be rejected.

The findings do not support the results of previous work which demonstrated that scientists are deemed more credible and trustworthy than other sources of information (Pew, 2005; Poortinga et al., 2004; Poortinga and Pidgeon, 2003; Leiserowitz et al. 2013).

In addition, the results are also not supportive of the previous literature which found that males were perceived as more credible than female counterparts (Armstrong and Nelson, 2005; Armstrong and McAdams, 2009; Steinpreis et al., 1999).

However previous literature has been mixed. Results of the current study show empirical support for White and Andsager (1991) who tested the effects of gender on perceived credibility on newspaper columnists and found that gender cues didn’t not have a significant effect.

In addition, a study by Burkhart and Sigelman (1990) found that participants, also university students, evaluated news articles similarly regardless of the gender of the author. This indicates that research into the relationship between gender stereotypes and perceived credibility could involve confounding variables that may influence results.

The results could imply the notion that society is moving towards equality. Perhaps times are changing, and maybe the lack of significant effects of gender and occupation suggests that the sample perceived men and women; and scientist and journalists as equally credible. It could be suggested that the sample did not possess schemas of traditional gender or occupation norms. A number of explanations may describe the patterns of results.

Firstly, there might have been increased awareness of societal inequalities due to recent media coverage of topics such as the gender pay gap that have challenged traditional norms.

Secondly, it could be noted that younger generations are seeing increasingly more women employed in typically ‘male’ professions (WISE, 2017) and are therefore socialised with different norms.

A shift towards equality is evident in the findings of previous literature. Miller at al. (2018) conducted a meta-analysis of 78 studies which looked at the proportion of gender stereotypes in the Draw-a-Scientist Test over the span of five decades. Although it was noted that children still associate science mostly with men, it was found that the proportion of female scientists being drawn by children has steadily increased over time. This could reflect a change of societal norms away from rigid gender stereotypical roles which might be used to explain the insignificant effects of gender and occupation cues in the present study.

On one hand, the findings of the experiment may indicate that the sample did not possess stereotypical gender or occupation norms. On the other hand, inadequacies in the experimental design to investigate the hypotheses may be responsible for finding no significant differences, when underlying stereotypes might have existed and were not portrayed in the results

Practical limitations

Characteristics of the sample population may have limited the findings of the present study.

Firstly, the majority of participants were female (88.2%). The gender distribution of participants may have biased the results because it is more unlikely that women will hold sexist stereotypes and therefore overall, men and women were perceived as equally credible regardless of their occupation and thus no significant differences were found between conditions. This explanation can be supported by the findings of previous literature which suggest that people are more likely to align their attitudes and rate the author of the source as more credible when they are perceived as similar to the self (Platow et al., 2000; White and Andsager, 1991). The majority of the sample were females studying a scientific subject. Considering that they defy traditional gender norms, it is unlikely that the majority of participants held stereotypical gender norms.

Subsequently, female participants who were assigned to either female condition might have strongly identified with the source and as a result the female authors were perceived as equally credible as male authors. This was reflected in the results as the Female Scientist was regarded as the most credible source, although the differences between conditions were not significant.

Had the gender of participants been evenly distributed, the results may have been skewed towards higher perceived credibility of male sources and more willingness to engage in environmentally friendly behaviour. A restriction of this explanation is that it cannot explain the non-significant difference of perceived credibility and behavioural intentions between journalist and scientist authors.

Furthermore, the results do not support previous research such as Armstrong and Nelson (2005) that have found significant effects of gender cues on source credibility despite the majority of participants being female (73%). Although the imbalance of gender distribution is likely to have influenced the non-significant differences, this explanation is incomplete because it cannot fully explain the absence of significant differences between experimental conditions.

A second limitation of the sample population was the average age of participants. All participants were aged between eighteen and twenty-four years and therefore it could be argued the sample was not representative of the population. Older age groups may have been socialised to have different norms, and therefore would possess different attitudes regarding gender roles than younger generations.

It could be implied that no significant differences were found because gender stereotypes might be absent in younger populations, indicating that society is moving towards equality. Furthermore, it is unlikely that participants should have the same societal norms as participants of previous studies from two or three decades ago, which would explain the lack of support for previous literature that found differences in trust and credibility in sources. However this presumption can only be inferred.

Another limitation of the sample population that might have influenced the findings to be non-significant was the motivations behind participation. All participants were undergraduates who took part in exchange for compulsory course credit. The main interest of the sample for completing the study could have been solely to get credit, forfeiting the interest and attention paid to the content of the study. For example, some participants completed the entire study in less than several minutes, making it extremely likely that they did not pay attention to the content.

This was reflected in the responses to the attention checks, of which nearly all participants gave at least one incorrect answer. It was critical that participants paid attention to cues in the present study so that the aims would have been tested as accurately as possible, therefore shallow processing of the cues would make any results redundant. However, some previous literature has argued that when little or no attention is exhibited, schemas are activated and relied upon to judge the credibility of the source (Petty et al., 1981; Cacioppo et al., 1986).

Hence it could be argued that despite somewhat lack of attention paid to the contents, it was likely that participants relied on the characteristics of the source to inform their opinion and yet still no significant differences were found. This indicates that there might be other explanations for the pattern of results such as lack of stereotypical gender and occupation schemas or perhaps participants did not fully process gender and occupation cues.

It could be argued that cues were not made sufficiently salient. As a result of this, participants’ behavioural intentions and perceived credibility of the source were not driven by the gender or occupation independent variable manipulations. To cue the gender and occupation of the author, one slide was presented to participants that explained who the author was and their place of work, and only a picture and a name was used at the top of the article. Perhaps the gender and occupation cues were not apparent enough to capture the attention of the participants and therefore any underlying schemas would not be activated. However results show that 97.5% of participants correctly answered the gender attention check so this explanation of the reason behind non-significant findings is incomplete.

In regards to the occupation cue however, a substantial proportion of participants failed to answer correctly (11.8%), therefore it could be argued that the occupation cues were not competently salient and therefore no attitudes towards occupation were activated which might have influenced perceived credibility and behavioural intention.

On the contrary, perhaps participants processed the cues, however the behavioural adaptions suggested by the author might have been considered as too difficult and thus no significant differences between conditions were found. The three behaviours were specifically selected to be included in the article because they were likely to have been considered as more challenging. Hence a ceiling effect was avoided that might have occurred had the article included behaviours that were perceived to be easy and therefore everyone would say they would be willing to do them.

It was predicted that the influence of the source would be more noticeable if behaviours that were challenging were used. Some habits, however, are notoriously hard to change.

The results of the current study support empirical findings from Whitmarsh et al. (2011) that showed the reluctance of the population to adapt eating and transport behaviours. 57.2% of participants claimed that they were never willing to avoid eating meat. In addition, 90% of people knew that aeroplane emissions were a cause of climate change, yet 35.4% were never willing to adapt behaviour. Perhaps it was too difficult for people to change these behaviours and therefore the investigation into the differences between the influences of each source would have been made redundant. In particular, an online article might not be enough to encourage change of behavioural intentions on these suggestions.

This may be a credible explanation for the lack of significant difference between the influences of the authors; people were no more willing to change behaviour regardless of how they perceived the source.

Furthermore, it could be noted that behavioural intentions are only a partial indicator of behaviour and therefore cannot be assumed as accurate representations of actual behaviour. On one hand, the Theory of Reasoned Action model (Fishbein and Ajzen, 1974) would suggest that intention is the strongest predictor of behaviour, supported by a meta-analysis of 128 studies, which concluded that environmental behaviour was most strongly predicted by verbal commitment (Hines et al., 1986).

The model states that intentions are exclusively predicted by attitudes and norms, consequently this supports the idea that the sample did not possess stereotypical attitudes and norms and therefore no significant differences of behavioural intentions were found between conditions.

On the other hand, previous research has shown a gap between intention and behaviour (La Pierre, 1934; Wicker, 1969), which discredits the Theory of Reasoned Action (Fishbein and Ajzen, 1974) as it falsely suggests that attitudes and norms are sole factors in predicting behaviour.

More advanced theories such as the Theory of Planned Behaviour (Azjen, 1991) limit the Theory of Reasoned Action as it acknowledges that there are other factors aside from attitudes and norms that might mediate behavioural intentions, such as perceived behavioural control. This is the idea that the extent to which people believe they can do an action should increase the likelihood of behaviour, which might justify the notion that the behaviours in the article were perceived to be too difficult and resulted in no significant differences between the effect of the source on behavioural intentions.

This limits the methodology of the present study as it discredits the validity of using behavioural intentions as a measure of attitudes and norms. Although attitudes and norms may have a significant influence on behavioural intention, it is likely that other factors such as perceived behavioural control prevented the effects of credibility of the source from being reflected in the intention.

Consequently, it could be argued that that to measure behavioural intentions is not a credible way of assessing attitudes, as there could have been a weak relationship between attitude and behaviour.

Perhaps the effects for each behavioural intention measure were too modest and combining all three behavioural intention measures would have produced noticeable differences between conditions. Four behavioural intentions were combined to form measures that were relevant to each behaviour to jointly predict the willingness, as stated in the Multiple Act Criterion; correlations are stronger when a series of behaviours are used to measure behaviour rather than using one behavioural example (Fishbein and Azjen, 1974).

However previous literature has shown a stronger correlation of attitudes and behaviour towards the environment when all measures are combined (Weigal and Newman, 1976). Therefore combing all three behavioural measures may have improved the measure and perhaps shown significant differences between conditions.

Implications for policy and practice

The findings of the current study provide implications for public policies which address climate change. The results might be useful for future communication of environmental issues, in particular; choosing the most efficient method in which to present information to the public. Perhaps an online article is not enough to encourage public acceptance of the message and subsequently willingness to adapt certain behaviours, and more effective methods of communication are needed to encourage the adjustment of challenging behaviours such as eating habits, aeroplane travel and switching energy providers.

The results may also imply that a key motivator behind environmentally friendly behavioural adaptions is not the characteristics of the source. This implies that for environmental communication to be effective, it is perhaps the content of message that would have an influential effect on evident limited motivation behind behavioural change. For example the result implicate that targeting perceived behavioural control of the public is likely to be influential.

Future research

In consideration of the practical limitations of the present study, there is an abundance of developments that could be used to improve the experimental design in future research to increase internal and external validity, as well as reliability.

Firstly, instead of using explicit measures to look at the differences between perceived credibility of sources, an implicit attitudes test (IAT) could be used to investigate true attitudes towards gender and occupation norms. There may be inconsistencies between private attitudes and public expression. The present study would have only assessed public expression of attitudes and therefore an IAT would investigate the true attitudes towards gender bias, reduce response bias and social desirability and therefore reliably test differences in perceived credibility between males and females, and scientists and journalists.

Future research should ensure there is an even distribution of males and females in the sample. Perhaps use a larger sample from the general public to get an even proportion of gender and a range of ages so that findings would be representative of the entire population. Older generations may be socialised with different norms therefore using the general public as a sample may be representative of the population and perhaps show gender and occupation schemas. Using the general public instead of mostly female undergraduates in a scientific course may be more representative as the niche sample in the present study may share attitudes towards occupation and gender norms.

Future research into this topic could include a prime to activate any existing gender or occupation stereotypes before the study. Low accessibility of stereotypes might have been present in the current study, resulting in no significant differences between conditions. Priming participant’s stereotypes would activate implicit attitudes and norms which may influence the perceived credibility of a source. This could be done by using the ‘Draw-a-Scientist Test’ used in previous literature (Chambers, 1983) or by adopting the method of Powlishta (2000) and presenting participants with pictures of men and women and asking them to rate the individual on masculine or feminine personality traits.

Conclusion

To conclude, the present study found no significant differences of perceived credibility between male and female, or scientist and journalist authors. Furthermore, neither one of the sources encouraged participants willingness to engage in the specific behaviours significantly more than the others. This might indicate a shift in the direction of equality in society or perhaps improvements of the design are required.

References

Aoyagi-Usui, M. (2008). An analysis of the effective factors for promoting pro-environmental actions from the information gain and social capital point of view. Review of Environmental Economics and Policy Studies, 1(2), 37-50.

Armstrong, C. L., & McAdams, M. J. (2009). Blogs of information: How gender cues and individual motivations influence perceptions of credibility. Journal of Computer-Mediated Communication, 14(3), 435-456.

Armstrong, C. L., & Nelson, M. R. (2005). How newspaper sources trigger gender stereotypes. Journalism & Mass Communication Quarterly, 82(4), 820-837.

Buldu, M. (2006). Young children’s perceptionsof scientists: a preliminary study. Educational Research, 48(1), 121-132.

Cacioppo, J. T., Petty, R. E., Kao, C. F., & Rodriguez, R. (1986). Central and peripheral routes to persuasion: An individual difference perspective. Journal of personality and social psychology, 51(5), 1032.

Cardiff University (2017). Gender Pay Gap Reporting. Retrieved from https://www.cardiff.ac.uk/__data/assets/pdf_file/0009/1138149/Cardiff-University-gender-pay-gap-report-2017.pdf

Doran, P. T., & Zimmerman, M. K. (2009). Examining the scientific consensus on climate change. Eos, Transactions American Geophysical Union, 90(3), 22-23.

Earle, T. C., & Cvetkovich, G. (1995). Social trust: Toward a cosmopolitan society. Greenwood Publishing Group.

Eurobarometer (2001). Eurobarometer 55.2 – Europeans, science and technology. Brussels: European Commission Directorate-General for Research.

GOV.UK (2017). Retrieved from https://gender-pay-gap.service.gov.uk/

Hansen, A. (2011). Communication, media and environment: Towards reconnecting research on the production, content and social implications of environmental communication. International Communication Gazette, 73(1-2), 7-25.

Hoban, T. J. (2004). Public attitudes towards agricultural biotechnology.

Hoppner, C., & Whitmarsh, L. (2011). Public engagement in climate action: policy and public expectations. Engaging the public with climate change. Behavior change and Communication, 47-65.

IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.

Johnson, B. B. (1999). Exploring dimensionality in the origins of hazard-related trust. Journal of Risk Research, 2(4), 325-354.

Konisky, D. M., Milyo, J., & Richardson, L. E. (2008). Environmental policy attitudes: Issues, geographical scale, and political trust. Social Science Quarterly, 89(5), 1066-1085.

Leiserowitz, A. A. (2005). American risk perceptions: Is climate change dangerous?. Risk analysis, 25(6), 1433-1442.

Leiserowitz, A. A., Maibach, E. W., Roser-Renouf, C., Smith, N., & Dawson, E. (2013). Climategate, public opinion, and the loss of trust. American behavioral scientist, 57(6), 818-837.

Lipset, S., & Schneider, W. (1983). The Decline of Confidence in American Institutions. Political Science Quarterly, 98(3), 379-402.

National Research Council (Ed.). (1989). Improving risk communication. National Academies.

Pearson, J. C. (1982). Gender, Similarity, and Source Credibility.

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.

Petty, R. E., Cacioppo, J. T., & Goldman, R. (1981). Personal involvement as a determinant of argument-based persuasion. Journal of personality and social psychology, 41(5), 847.

Pew Research Center for the People and the Press. (2005). Public more critical of press, but goodwill persists. Retrieved from http://people-press.org/reports/display.php3?ReportID=248.

Poortinga, W., Bickerstaff, K., Langford, I., Niewöhner, J., & Pidgeon, N. (2004). The British 2001 foot and mouth crisis: a comparative study of public risk perceptions, trust and beliefs about government policy in two communities. Journal of Risk Research, 7(1), 73-90.

Poortinga, W., & Pidgeon, N. (2003). Public Perceptions of Risk, Science and Governance: Main findings of a British survey of five risk cases. Norwich: University of East Anglia and MORI.

Renn, O., and Levine, D. (1991). Credibility and trust in risk communication. In R. E. Kasperson, and P. J. M. Stallen (Ed.), Communicating risks to the public (pp. 175-218). Dordrecht: Kluwer Academic.

Ross, K., & Carter, C. (2011). Women and news: A long and winding road. Media, Culture & Society, 33(8), 1148-1165.

Sandvik, H. (2008). Public concern over global warming correlates negatively with national wealth. Climatic Change, 90(3), 333-341.

Slovic, P. (1993). Perceived risk, trust, and democracy. Risk analysis, 13(6), 675-682.

Steinpreis, R. E., Anders, K. A., & Ritzke, D. (1999). The impact of gender on the review of the curricula vitae of job applicants and tenure candidates: A national empirical study. Sex roles, 41(7-8), 509-528.

The data for this paper was generated using Qualtrics software, Version 2018 of Qualtrics. Copyright © 2018 Qualtrics. Qualtrics and all other Qualtrics product or service names are registered trademarks or trademarks of Qualtrics, Provo, UT, USA. https://www.qualtrics.com.

The Equality Act 2010 (Gender Pay Gap Information) Regulations 2017. Retrieved from https://www.legislation.gov.uk/ukdsi/2017/9780111152010

Whitmarsh, L. E. (2005). A study of public understanding of and response to climate change in the South of England (Doctoral dissertation, University of Bath).

Whitmarsh, L. (2009). Behavioural responses to climate change: Asymmetry of intentions and impacts. Journal of environmental psychology, 29(1), 13-23.

Whitmarsh, L. (2011). Scepticism and uncertainty about climate change: Dimensions, determinants and change over time. Global environmental change, 21(2), 690-700.

Whitmarsh, L., O’Neill, S., Seyfang, G., & Lorenzoni, I. (2009). Carbon Capability: what does it mean, how prevalent is it, and how can we promote it. Tyndall Centre for Climate Change Research, Norwich.

Whitmarsh, L., Seyfang, G., & O’Neill, S. (2011). Public engagement with carbon and climate change: to what extent is the public ‘carbon capable’?. Global environmental change, 21(1), 56-65.

WISE Campaign (2017). Women in STEM workforce 2017. Retrieved from https://www.wisecampaign.org.uk/resources/2017/10/women-in-stem-workforce-2017

WISE Campaign (2017). The STEM Education Pipeline. Retrieved from https://www.wisecampaign.org.uk/resources/2017/11/from-classroom-to-boardroom-the-stem-pipeline-2017

Ziman, J. M. (1969). Information, communication, knowledge. Nature, 224(5217), 318.