Is public engagement gendered? An analytical proposal using some evidence from Italy

Abstract

The relationship between gender and public engagement in the literature is unclear. Unlike with other factors, research has not conclusively shown whether gender affects involvement in public engagement activities and, if it does so, in what direction. This article aims to investigate this ambiguity and intends to contribute to the understanding of underlying factors. Using survey data from a large national sample of Italian academics from all disciplines (N = 5123), we find a gender gap in public engagement activities in the mass media and not in community-based activities. These results suggest a different analytical approach for investigating gender differences in public engagement.

Keywords

gender differences mass media public engagement science in society

Over the past 30 years, the relationship between researchers and society has been the subject of increasing attention from the institutions that govern and fund research (Burchell, 2015). Since the Royal Society’s encouragement (1985) for scientists to consider communication with the non-expert public a duty and for science to leave its ivory tower to enter the public space (Gibbons, 1999), universities and academics have been called upon to rethink their relationship with society.

The extent of some changes that have occurred over the last years has led to the emergence of new needs related to expert knowledge, scientific culture and public trust, calling for new attention regarding the implications of the ties between science and society. The complexity of technological and scientific developments, the new opportunities for individuals to find every kind of information and the quantity and variety of this, the dis-intermediation of communication and information as well as the social and political consequences of these changes, make the issues of citizens’ scientific literacy and support for science particularly urgent (Nichols, 2017). The concern of the diffusion of (broadly understood) scientific culture within society calls into question the role of the university as the main place for the production and reproduction of scientific knowledge, and in particular, the commitment of academics in science communication activities towards different audiences. Scientists are therefore called upon in first person to engage with society, to interact, connect and facilitate the relationship between the scientific world and the world of non-experts. In this context, the importance of the third mission of university has developed as a critical dimension in university activities, and particularly the relevance of public engagement of academics. Indeed, among the third mission’s dimensions (Schoen and Theves, 2006), public engagement has perhaps been the one to produce the most reflection, evolving over time and varying its definitions and activities included. The redirection of agenda from a one-way communication perspective – focused on the deficit of knowledge and the need for public education – towards a two-way dialogue – focused on public trust and the need for involvement of the public in citizen science projects – has implied an extension of activities: in addition to the outreach activities (such as media work, public lectures and debates, science festivals, work with museums and schools, writing for lay audiences and open lab events), other activities have emerged, in which scientists engage with society in the early phase of research and not only to communicate results (Burchell et al., 2017). In the United Kingdom, however, where the focus on public engagement agenda dates back to the 1980s, the relationship with mass media and policy-making is not included among public engagement activities (Research Councils UK, 2010). On the other hand, in Italy, the National Research Evaluation Agency (ANVUR) proposed 16 activities, including communication with the public through mass media and participation in policy-making (ANVUR, 2015: 41), a move echoed by researchers such as Hamlyn et al. (2015).

This article focuses on the public engagement activities of Italian academics, sometimes called ‘popularization activities’, ‘science communication activities’ or ‘public outreach activities’, designed for and directed at the non-academic public, composed by diverse and non-specialized audiences. We aim, in particular, to provide empirical evidence about the participation of women scientists in these activities and shed light on gender issues in public engagement, since in the literature this is unclear. What is the involvement of women in these activities? Is it different in terms of quantity and quality, from that of men?

This issue is relevant for at least two reasons. Gender inequalities in higher education, and particularly in science, are well-known and documented (European Commission, 2019). Given the importance of engagement with society, participation in these activities could be, in the near future, evaluated as one of the criteria for academic career prospects, thus representing an issue that may reduce or intensify existing inequalities for women. Furthermore, a greater presence of women in science can be achieved over time through a wider choice of scientific courses in higher education (and of a scientific career) by girls, who still today tend to be segregated within humanities and social sciences (Barone, 2011). Greater involvement of women in public engagement activities could contribute to feminize the image of science and to bring different role models for girls and increase the likelihood of choosing a scientific course of study, decreasing gender segregation in these fields.

1. Public engagement and associated factors

Scholars have identified some factors that can encourage or hinder participation in public engagement activities. Among these, the gender factor is the most ambiguous because it is not clear if it discriminates the participation in these activities, and if so, in which direction. In fact, in some studies, being male is positively associated with the engagement (Bentley and Kyvik, 2011; Besley et al., 2012; Crettaz von Roten, 2011; Kreimer et al., 2011), in others, it is negatively associated (Jensen, 2011; Jensen et al., 2008), and in others, it is not associated (Anzivino et al., 2018; Dudo, 2012). This ambiguity does not exist for the other personal and academic characteristics. Age is usually positively associated with public engagement, as well as academic rank, seniority and belonging to some academic fields. In fact, being aged, senior or full professor increases the probability of being involved in public engagement activities (Anzivino et al., 2018; Bentley and Kyvik, 2011; Jensen, 2011; Jensen et al., 2008; Kyvik, 2005; Scamuzzi and Tipaldo, 2015; Thune et al., 2016), and all studies show that social science and humanities’ scholars – when these disciplines are included – are more involved in public engagement activities than those belonging to STEM disciplines (Anzivino et al., 2018; Burchell et al., 2017; Jensen, 2011; Kreimer et al., 2011; Kyvik, 2005; Rainie et al., 2015; Thune et al., 2016).

In this study, we aim to address the ambiguity of results about the relationship between gender and public engagement, supposing that some methodological elements with which scholars study public engagement are at the basis of the conflicting results in the literature, namely, the variety of indicators used in the analysis and how these indicators are combined. After a review of these studies, we will provide some empirical evidence on the Italian case to support this hypothesis.

2. Public engagement studies and gender: Methodological issues

As Bauer and Jensen have previously stated (2011), comparing public engagement studies is not straightforward. They differ with regard to data collection, the definition of the populations and questions asked. Some studies are based on institutional and administrative data, others on qualitative interviews and others on questionnaires. Moreover, some studies involve scientists from all disciplines – including social sciences and humanities – others involve only some, more frequently STEM disciplines. As a result, the measurement of academics’ involvement in public engagement activities is very differentiated. Some studies ask open questions, others for a list of activities engaged in over the last 12 months or in the last 3 or 5 years with dichotomous answers, others using rating scales such as ‘no activity, some activity, much activities’, or ‘never’, ‘rarely’, ‘often’, ‘very often’, others for the number of times in which each activity was performed, and so on.

In addition to these remarks by Bauer and Jensen (2011) related to data collection, we would highlight the importance – especially for studying the gender issue in public engagement – of the type of activities included in the study, and the data management after data are collected. In particular, we think that the choice of the activities on the list in which we ask respondents about their engagement, how the engagement is measured and how the activities are combined in the analysis, could determine the ambiguity of the role of gender in literature.

Usually, scholars compute only one general index of public engagement, sometimes combining a single dichotomous variable that indicates the participation in at least one activity; sometimes adding dichotomous variables in a single variable that indicates the number of activities performed; sometimes adding scale-variables in a single variable that takes into account the variety and the frequency of activities performed. This range of possibilities in the composition of a public engagement dependent variable could be one of the reasons why the gender effect is unclear.

Looking at some international studies about public engagement, we can recognize a trend: studies that include a large number of mass media activities – proportionally – out of the total of public engagement activities considered, show that men are more engaged than women. Those that include more community-based activities and few or no mass media activities show that women are more engaged than men or do not find statistical gender differences.

Starting from this last group of studies, in the work of Thune et al. (2016), a dichotomous variable of public engagement (at least one of the dissemination activities performed) is used. It includes just one media activity – publication of contributions in the popular press – and three other activities – contributions to public debate such as participation of academics in debates, the participation in meetings/conferences with users/practitioners, participation in lectures/talks to users/practitioners – and men and women result as being equally engaged. Also, the Dudo (2012) study on biomedical researchers shows no differences between males and females in public engagement. Public engagement is computed as a single variable, where the community-based activities (nine activities) are more than twice the activities through mass media (four activities).

Other studies show that women are more likely to be involved than men. Jensen (2011) in an extensive survey on all disciplines using multivariate analysis, finds that French women scientists are more active in public engagement. The indicators of public engagement in this study are participation in conferences for a general audience, in exhibitions, in open doors, actions taken to help associations in understanding scientific aspects of their activity, actions taking place in schools, publications of books/CD-ROM for the general public, activities in the press, participation in radio/television/movies and popularization sites on the Web. A single dichotomous variable synthesizes these indicators: the academic has participated in at least one activity or not. By this procedure, mass media activities account for just over a fifth (22%; two media activities – press and radio/TV/movies are susceptible to be affected by the gender gap on the total of nine activities). Johnson et al. (2014), in their qualitative study, find that women scientists are more involved in public engagement than men, but the activities which are more frequently performed are those with children and school and public lessons, whereas media activities are less. Moreover, female scientists are more interested and motivated by the goal of increasing the participation of women in science, then those activities directed at students.

On the contrary, other studies indicate a greater engagement of men. Bentley and Kyvik (2011) show that men publish more than women in newspapers and magazines, and popularization is measured by a dichotomous variable where the two categories are ‘at least one popular publication’ and ‘no popular publications’. Also, in the work of Kreimer et al. (2011), males are more engaged than females. They consider 17 popularization activities and over half are media communication activities (i.e. TV or radio host or panellist; radio, newspaper, television, magazine, websites interviews; magazine, newspaper and websites articles). The dependent variable of public engagement is a synthesis of all activities, and it is dichotomous: participation in at least one activity or not in the last year. In their conclusion, the authors highlight the different character of the activities, some more prestigious – interviews on television and newspapers – than others – open doors or school talks – due to the role played by journalists who more frequently involve ‘experts’ and consequently the highest-ranking researchers. This kind of consideration could be extended to the gender issue since the image of expertise is often affected by stereotypes that include gender. On the other hand, authors in this study highlight how academics who did not engage in any activity, declared that if they had been offered the opportunity, they would have participated. This according to the authors underlines how engagement is not only an individual initiative, and it is also necessary to take into account the role of journalists who are looking for some researchers more frequently and the institutional role in promoting these activities.

Besley et al. (2012) provide another crucial empirical indication in the direction of the necessity to take into account the diverse character of the activities. When they use, in their analysis, two indicators – one for media communication activities (how often researchers talk with journalists about research results) and one for other communication activities (how often researchers talk with the general public about science or research results) – gender differences (male are more engaged) are shown for the media activities and not for others. When they use only one index for public engagement, where three media activities are contemplated out of four, males are more engaged than females, even though women consider engagement to be more important than males.

Moreover, Crettaz von Roten (2011) shows that men are more engaged than women, even though the attitudes towards public outreach and engagement activities are the same. She considers 17 activities, of which four are media activities. Although the number of media activities is not proportionally high, she computes a synthetical measure of public engagement, adding scores on the 17 items. This kind of computing allows researchers to take into account the number of activities performed and their frequency. Therefore, if women are less involved than men in media communication activities, they will result as being less engaged. This consideration has to be read in the light of her analysis of media contacts of researchers. In fact, she finds that the media contact women less often than men, providing interesting elements for rethinking of a different conceptualization of public engagement to analyse if and how men and women differ.

3. Hypotheses

Following this analysis of empirical results from international literature, and considering the literature that does not indicate any differences in attitudes towards public engagement for men and women (Crettaz von Roten, 2011; Poliakoff and Webb, 2007), we hypothesize that some activities included in the definition of public engagement are affected by the gender gap more than others, namely, activities in the mass media; and that by keeping separate activities in the media from the other activities, gender role could be more clarified.

In particular, we expect there to be, both in the concept of public engagement as well as in the answers of respondents, two distinct dimensions, one for mass media activities and one for community-based activities (Hypothesis 1).

Moreover, since the visibility of scientists in the media tends to reproduce the structure of inequalities existing in the scientific community (Bucchi, 2002); the glass-ceiling phenomenon (European Commission, 2019), in which women are proportionally less represented in apical career roles (in this case, in the position of full professor, department director and so on); the so-called Matthew effect (Merton, 1986), through which well-known scientists enjoy more trust and credit, and the parallel Matilda effect (Rossiter, 1993), through which women in science are systematically underestimated in their work; and the amount of media contacts is positively associated with rank and research productivity (Peters et al., 2008) and to be male (Crettaz von Roten, 2011), we expect women will be less engaged in media communication activities (Hypothesis 2).

At the same time, since community-based engagement or direct engagement is considered less prestigious as it is not a source of visibility and the involvement of academics does not depend on the journalists’ invitation (Kreimer et al., 2011), we do not expect to see gender differences (Hypothesis 3). Finally, we hypothesize that the attitudes towards engagement do not affect the lesser participation of women in media activities (Hypothesis 4).

4. Data and methods

Data

Data used in this article have been collected through a national survey on academics’ third mission activities, carried out between the end of 2015 and the beginning of 2016. It has been possible to collect information from 5123 respondents working at 62 universities, with a response rate of 34.2%, on 27 different third mission activities. The sample was randomly selected from the Ministry of Education lists, according to two stratification criteria: the field of teaching and the university’s geographical location within a macro-region. The field of teaching included seven categories: Humanities and arts, engineering and architecture, social and behavioural sciences, business, economics and law, mathematics, physical and life sciences, agriculture and veterinary, and health. The university’s geographical location included five categories, corresponding to the traditional division of the country in macro-regions: North-West, North-East, Centre, South and the Islands. The sample is representative of the entire population of Italian academics who work in a public university.¹ In addition to being a probabilistic sample, the distribution of the two stratification variables – field of teaching and geographical area – and the distribution of the other relevant variables – gender and academic position – are the same in the final sample and population.

Measures

Public engagement activities: Among the 27 third mission activities investigated by the questionnaire through a Likert-type scale, some were related to public engagement. We have selected five activities that are characterized by being addressed to a general and non-expert public. Three of these activities were addressed to the territory in which the university is located: collaboration in the realization of cultural or recreational or sporting events (exhibitions, museums, concerts, scientific dissemination festivals, etc.); participation in projects addressed to primary and/or secondary schools; participation in conferences, meetings, training activities addressed to the general public. The other two activities were explicitly addressed as scientific dissemination in the mass media: scientific dissemination through mass media interventions (press, radio, TV, Internet, digital publishing, social media and blog); contributions to the public debate through mass media interventions (press, radio, TV, Internet, digital publishing, social media and blog). Participation in each of these activities referred to the 5 years prior to the interview and the frequency was observed on a four category scale: never, rarely, quite often and very often.

Independent and control variables: Gender is our independent variable and it was collected at the start of the questionnaire, with 4.2% of missing data, which we have excluded from the analysis, leaving 56.4% of men and 39.4% of women. As control variables in the multivariate analysis we have included age in years; academic position (full professor, associate professor and assistant professor); the field of teaching and the geographical area where the university is located (as considered during the sampling); university size, classified on the basis of the number of students (small, up to 10,000 students; medium, from 10,001 to 20,000 students; large, from 20,001 to 40,000 students and mega, more than 40,000 students); academic productivity, measured by the number of articles or chapters published in academic journals or books, the number of scientific books authored and the number of scientific books edited. Attitudes towards engagement are indicated by the answers on a Likert-type scale (completely disagree, disagree, agree and completely agree) to the following question: ‘The following sentences concern the relationship between university and external organizations. Please, value your degree of agreement or disagreement with each of them: Universities should increase their social relevance’. This indicator has been inserted in the model as dichotomous, considering together ‘totally disagree’ (.5%), ‘disagree’ (4.6%) and ‘agree’ (38.3%) and separately the answers ‘totally agree’ (56.5%); this kind of recoding was necessary due to the low discriminatory power of this indicator.

Techniques

To answer our research questions, we relied on several statistical techniques. We used bivariate analysis to investigate the extent of public engagement activities for men and women, and factor analysis to test the multidimensional hypothesis, that is the existence of different latent dimensions in the concept of public engagement as it results from respondents’ activities. Finally, to test whether men and women differ in engagement with society in each dimension of public engagement, also considering the academic position, age, discipline and other factors indicated by literature as potentially relevant, we relied on regression techniques, namely stepwise linear regression. This technique allows us to use all the available information estimating the effect of gender, net of other variables and to look at the variation of the parameters when we include the control variables. In the supplemental material, we add the logistic regression models that we developed with the aim of controlling the results obtained by the linear regression models. The reason for relying on new models using different techniques are twofold: on one hand, if the distribution of our dependent variables (indexes of public engagement, constructed after factor analysis) is biased, then we control the association between the different dimensions of public engagement and gender by a logit function, to be sure that results are not due to the initial bias; on the other hand, the technique most frequently used in the literature is the logistic regression, since scholars usually compute a single dichotomous dependent variable of public engagement, so we also offer results in the most common format.

5. Findings

Results from bivariate analysis (Table 1) show no gender differences for two of the three activities addressed to the territory where the university is located, a small difference for activities directed at schools and significant differences for the two activities of dissemination through the mass media.

Table 1.

Public engagement activities by gender (%; men = 2822; women = 1955).

		Men	Women
Carrying out of sport, leisure or cultural events	Never	43.8	43.1
	Rarely	26.2	27.6
	Quite often	21.4	21.4
	Very often	8.6	7.9
	Total	100	100
	Chi-square Pearson = 1.695, sig. = .638
Projects with primary or secondary schools	Never	43.3	39.2
	Rarely	30.4	31.0
	Quite often	19.9	21.5
	Very often	6.4	8.2
	Total	100	100
	Chi-square Pearson = 11.835, sig. = .008
Meetings, conferences or training activities	Never	14.4	15.2
	Rarely	27.2	28.5
	Quite often	36.9	35.8
	Very often	21.6	20.5
	Total	100	100
	Chi-square Pearson = 2.264, sig. = .519
Scientific dissemination through the mass media	Never	41.1	50.4
	Rarely	41.0	36.7
	Quite often	15.0	10.8
	Very often	2.9	2.0
	Total	100	100
	Chi-square Pearson = 46.322, sig. = <.001
Contribution to public debates through the mass media	Never	51.7	63.0
	Rarely	34.5	27.0
	Quite often	11.0	8.5
	Very often	2.8	1.4
	Total	100	100
	Chi-square Pearson = 64.837, sig. = <.001

For both men and women, the most frequent activity is the participation in conferences, meetings and training courses for the general public. There are no relevant differences in the intensity of engagement, except for a slightly higher propensity for women to participate in projects with primary or secondary schools (statistically significant). Relating to public engagement activities in the mass media, the differences between men and women are more substantial. The frequency of engagement is higher for men (+9% for scientific dissemination activities, and +11% for contribution in public debate), as well as the intensity that, although generally low, is higher than for women.

By factor analysis,² we tested the existence of two different dimensions of public engagement, in the respondents’ answers. Results (Table 2) show that the first component is mainly related to the two activities of science communication through the mass media and the second component is shaped by the three activities carried out within the territory. We have named ‘Media engagement’ the first component and ‘Local engagement’ the second component. These two dimensions of public engagement are different on several levels. Local engagement refers to the communication activities that imply a direct interaction with the public within the local community. Media engagement refers to communication activities directed at a potentially larger and general public that implies an indirect relationship with them. Moreover, as we have already seen in the literature review, these two dimensions differ depending on the origin of the initiative for public engagement activities. For media engagement, the initiative is external, it comes from editorial staff, whereas for local engagement it is more likely to come from the academics or the department.

Table 2.

Factor analysis on public engagement activities – rotated component matrix, varimax rotation.

	Component 1	Component 2
Carrying out of sport, leisure or cultural events	0.393	0.704
Projects with primary or secondary schools	0.004	0.865
Meetings, conferences or training activities	0.414	0.684
Scientific dissemination through the mass media	0.892	0.221
Contribution to public debates through the mass media	0.903	0.183

To understand the gender effect on involvement in these two dimensions of public engagement, we calculated two indexes on the basis of the factor analysis results, adding the scores for each item and dividing the total by the number of items³ composing the indexes. We used them as dependent variables in two linear regression models: one for the index of local engagement, one for the index of media engagement (both varying from 1 – not engaged – to 10 – maximally engaged). This computation allows us to take into account, at the same time, the variety of activities performed and the frequency of engagement. Moreover, we calculated, by the same procedure, a third dependent variable that considers all activities of public engagement together in a single index and we used it in another linear regression model. Preliminary results from the bivariate analysis on the three indexes confirm what we have observed looking at the single activities and gender. Local engagement does not differ between men (M = 4.5, SD = 2.3) and women (M = 4.6, SD = 2.4), while media engagement does (men: M = 3.2, SD = 2.2; women: M = 2.7, SD = 2.0). Public engagement considered as a whole differs significantly but slightly, incorporating these two trends, therefore men as a result tend to be more engaged than women (men: M = 4.0, SD = 2.0; women: M = 3.8, SD = 2.0).

The multivariate analysis (Table 3) offers a deeper picture. Results from the first model (local engagement as dependent variable) confirm the absence of substantial differences between men and women that we have observed in the bivariate analysis. Even considering the control variables, individual and academic factors – such as age, position, the discipline of teaching and academic productivity – and the institutional variables – size and macro-region in which the university is located – women and men are equal for the intensity of public engagement within the territory (Model I, Step I). On the contrary, results from the second model (media engagement as dependent variable) show that the gender effect on public engagement through mass media remains significant and relevant also controlling for the other variables: age, position, discipline, productivity, size and location of the university being equal, men have on average a higher score of 0.298 points of media engagement (Model II, Step I).

Table 3.

Linear regression model on local engagement (N = 4308); linear regression model on media engagement (N = 4303); linear regression model on public engagement (N = 4297).

	Model I		Model II		Model III
	Local engagement		Media engagement		Public engagement
	Step I	Step II	Step I	Step II	Step I	Step II
	Coeff. B	Coeff. B	Coeff. B	Coeff. B	Coeff. B	Coeff. B.
Intercept	4.206***	3.829***	1.537***	1.319***	3.142***	2.825***
Gender
Male	−.121	−.116	.298***	.301***	.048	.053
Female (Ref.)	0	0	0	0	0	0
Age (years)	−.003	−.004	.014**	.014**	.004	.003
Academic position
Full professor	.475***	.476***	.679***	.680***	.557***	.557***
Associate professor	.265**	.276**	.237**	.243**	.255***	.264***
Assistant professor (Ref.)	0	0	0	0	0	0
Discipline
Humanities and arts	.956***	.964***	.047	.051	.591***	.598***
Social and behavioural sciences	1.200***	1.174***	.813***	.797***	1.042***	1.020***
Business, economics and law	−.050	.008	−.193	−160	−.105	−.056
Mathematics, physical and life sciences	.041	.128	−.757***	−.707***	−.279**	−.206*
Architecture and engineering	−.040	−.039	−.457***	−.412**	−.209	−.143
Agriculture and veterinary	.084	.145	−.003	.030	.053	.102
Health (Ref.)	0	0	0	0	0	0
Academic productivity
No. of books published in 5 years	.096***	.094***	.125***	.124***	.108***	.106***
No. of edited/co-edited books published in 5 years	.126***	.121***	.138***	.135***	.130***	.126***
No. of articles published in 5 years	.006***	.006***	.006***	.006***	.006***	.006***
University location
North West	−.448***	−.426**	.004	.016	.–263*	.–245*
North East	−.328*	−.307*	.035	.046	−.177	−.161
Centre	−.656***	−.641***	−.066	−.058	−.415***	−.403***
South	−.242	−.254*	.123	.116	−.092	−.103
Islands (Ref.)	0	0	0		0	0
University size
Small	.957***	.950***	.575***	.571***	.797***	.790***
Medium	.373***	.379***	.263**	.266**	.327***	.332***
Large	.065	.073	.185*	.190*	.115	.122
Mega (Ref.)	0	0	0	0	0	0
Attitudes
Totally agree with the need for an increasing social relevance of university		.674***		.391***		.564***
Not totally agree with the need for an increasing social relevance of university (Ref.)		0		0		0
Adjusted R²	.096	.116	.149	.157	.134	.153

p < .05; **p < .01; ***p < .001.

It must be highlighted that the differences between the two types of engagement involve other aspects. Those that are interesting for our aims are related to the mass media logic and the journalists’ selection criteria of academics to involve in debates, speeches or interviews. Indeed, the factors relating to the Matthew effect and the scientist’s visibility – age, position and scientific productivity – matter more (in terms of significance and coefficient magnitude) for media engagement than for local engagement, for which, on the contrary, contextual factors count more. Particularly interesting also seems to be the fact that the publishing products circulating almost entirely within the scientific community (such as scientific articles) count equally for both local and media engagement but publishing activity accessible also to the general public (books), matters more for the media engagement.

Looking at the model with public engagement as the dependent variable (Table 3, Model III), we can observe how the slight gender difference existing at bivariate level, disappears by entering the control variables in the model (Step I): the initial greater engagement of males has likely been an effect of the academic rank or academic productivity, two factors associated with both engagement and gender. This result confirms that the conceptualization of public engagement as a whole, regardless of its different components, could mask the gender effect.

Since most of the studies in the literature compute public engagement as a dichotomous variable and test differences through the logit model, we repeated the analysis calculating the dependent variables as three dichotomies: at least one activity of local engagement (or of media engagement; or of public engagement) performed at least rarely; and no activity performed (never performed in the last 5 years). The logistic models (with the same control variables)⁴ produce the same results as the linear models regarding the role of gender: the gender effect is significant only for media engagement, net of academics’ other characteristics and contextual factors.

To test the fourth hypothesis, we carried out the same models on media engagement, local engagement and public engagement as a whole, entering the attitudes towards the role of the university in society. The gender regression’s coefficients remain essentially the same for the three models (Step II). Particularly for media engagement, attitudes do not reduce the gender difference, that is the attitudes (and the other factors such as age, rank, discipline, research productivity, university size and macro-regional context) being equal, males’ media engagement is higher than females.

6. Conclusion

This article aimed to show that the ambiguity in the literature concerning gender differences in public engagement could be ascribed to the great variety of activities included in the definition and analysis of public engagement, and specifically to the one-dimensional conceptualization of public engagement that does not take into account the diverse character of activities. Our results offered empirical evidence in this direction.

Results of the factor analysis confirmed the existence, also in the answers of the respondents, of two dimensions of public engagement – one more local and direct to the public, and one mediated and oriented to diverse and larger publics, through the mass media.

The multivariate analysis – carried out on the two dimensions of public engagement and public engagement as a whole – showed that studying the relationship between gender and public engagement, without taking into account the multidimensionality of the concept, could be misleading. Indeed, synthesizing public engagement activities in a single variable showed that gender does not differentiate the engagement. Analysing the relationship with the two dimensions of public engagement separately, showed instead that, other factors being equal, women do the same local engagement activities but fewer media engagement activities than men.

This difference in media engagement cannot be due to diverse attitudes towards the role of the university in the society, since it was statistically significant also controlled for these. Indeed, although both women and men think that universities should increase their social relevance, women are less engaged in mass media activities.

Different explanations could be proposed for these results, some exogenous to the scientific community and others endogenous. It is possible to assume that women, being more penalized in the academic career in terms of advancement and research productivity, are more concerned about the professional stigma associated with scientific dissemination activities or about the time to devote to communication which takes time away from research. But this kind of motivation would apply both to local and media engagement. Considering that, other factors being equal, the inequalities of participation was solely on the mass media activities dimension of public engagement, we believe the explanation would be inherent in the specific logic of mass media, which can cross paths with some endogenous aspects related to the unequal distribution of resources in academia that could make an academic more visible.

In our research, we do not have data on contacts between academics and journalists. However, from other surveys (Peters et al., 2008), also for the Italian case (Bucchi and Saracino, 2012), we know that the frequency of contact with the media is associated with the scientist’s position and with academic productivity – two factors on which women are strongly disadvantaged (Dubois-Shaik and Fusulier, 2015; Van Der Besselaar and Sandstrom, 2016) – and that contacts with media are more frequent for men, independently of position, age and faculty (Crettaz von Roten, 2011). Certainly, visibility is only one of the criteria for the selection of scientists by journalists, as is necessarily the topic of research and communication. The more the topic directly affects humans and their life and health, the more interesting is for the media (Summ and Volpers, 2016).

Gender seems to be an element crossing different dimensions of the media logic – that favours high visibility, scientific reputation, some disciplines, as the media engagement model seems to indicate. However, since our results show an autonomous impact of gender on media engagement, we can assume there is also an acceptance of a conventional idea of science in mass media logic that leads to the adoption of stereotyped communication models, whereby the image of a scientist is one of being older, with a high position, elevated scientific reputation and male. The literature on gender differences in mass media seems to support this hypothesis. Some studies show how some journalistic routines favour the selection of men rather than women as experts and guests. Time and priority constraints induce journalists to contact people whom they worked with in the past, whose communication style they know, whose characteristics are confidence and authority, thus reproducing the existing imbalance (Howell and Singer, 2017; Niemi and Pitkänen, 2017). Moreover, some media practices, such as the emphasis on private life (Mitchell and McKinnon, 2019), as well as the physical and aesthetical aspects (Kitzinger et al., 2008) of women when they are involved as experts, could be at the basis of their reluctance to appear in mass media due to the fear of being judged for work-unrelated reasons (Howell and Singer, 2017).

The emphasis on the role of mass media is necessary because of the consequences that the exclusion of women from scientific media communication could have. On one hand, on their academic career; and on the other hand, on the participation of young women in scientific educational courses and on their professional options. On the first point, having fewer opportunities in media engagement when public engagement is becoming strongly advocated at different levels (national and European) and, for this reason, having the potential in the near future to be among the evaluation criteria for the academic career, means for women to be disadvantaged in career paths where they have already been disadvantaged. On the second point, although there is a growing importance of direct communication of scientists through web channels, the role of journalists for mediating the relationship with the public is still important (Peters, 2014), not only because they contribute to making the scientific content more accessible for the public, but also because they contribute to constructing the image of experts and science, specifically in the traditional media (TV, press and radio). A larger presence of women who speak about scientific research in the mass media could contribute to making the image of science more feminine and to making it closer to young women, expanding their educational and professional perspectives towards scientific and technological fields, which at present are still affected by segregation and considered by many as male fields.

Finally, we have to remember the role of the university in promoting public engagement of their academics, particularly in the Italian context. Each university has been compelled to reflect on the third mission (of which public engagement is a part) and to provide incentives to academics for participating in activities that are becoming part of the academic profession, in addition to the traditional academic activities, teaching and research. Considering this, universities should promote women’s media engagement, in particular. Rather than general incentives, universities should focus on actions to encourage the participation of women in programmes or interviews in the press, television and radio, supporting them in specific training to build up their necessary skills to face the media logic and addressing their press officers to promote women as experts, particularly in those fields which are still perceived as male domains.

Supplemental Material

sj-pdf-1-pus-10.1177_09636625211002060 – Supplemental material for Is public engagement gendered? An analytical proposal using some evidence from Italy

Supplemental material, sj-pdf-1-pus-10.1177_09636625211002060 for Is public engagement gendered? An analytical proposal using some evidence from Italy by Monia Anzivino in Public Understanding of Science

Footnotes

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article: Data on Italian academics’ Public Engagement have been collected by the survey “Le attività di terza missione degli accademici italiani” within the research project “Universities, innovation and regional economies” co-funded by the Italian Ministry of University and Research and four Universities: Turin, Milan, Pavia and Florence (MIUR—Programmi di Ricerca Scientifica di Rilevante Interesse Nazionale 2010–2011, Contract No. 2010WN5KCP_004).

ORCID iD

Monia Anzivino

Supplemental material

Supplemental material for this article is available online.

Notes

Author biography

Monia Anzivino, PhD in Methodology of social research, is a research fellow at the University of Turin and is a member of the Centre for Study and Research on Higher Education Systems (CIRSIS), University of Pavia. Her research interests are related to higher education issues: public engagement and the role of scientists in society, the gender gap in the academic career, students’ performance and career. She also is interested in media effects on citizens’ attitudes towards politics and science.

References

Agenzia Nazionale di Valutazione del Sistema Universitario e della Ricerca (ANVUR) (2015) La valutazione della terza missione nelle università italiane. Manuale per la valutazione [The evaluation of third mission in the Italian universities. Handbook for the evaluation]. Roma: ANVUR.

Agenzia Nazionale di Valutazione del Sistema Universitario e della Ricerca (ANVUR) (2016) Rapporto biennale sullo stato del sistema universitario e della ricerca [Biennial report on the state of the university system and research]. Roma: ANVUR.

Anzivino

Ceravolo

Rostan

(2018) L’impegno pubblico e sociale [Public and social engagement]. In Perulli

Ramella

Rostan

e Semenza

(eds) La terza missione degli accademici italiani [The third mission of the Italian academics]. Bologna: Il Mulino, pp. 149–174.

Barone

(2011) Some things never change: Gender segregation in higher education across eight nations and three decades. Sociology of Education 84(2): 157–176.

Bauer

Jensen

(2011) The mobilization of scientists for public engagement. Public Understanding of Science 20(1): 3–11.

Bentley

Kyvik

(2011) Academic staff and public communication: A survey of popular science publishing across 13 countries. Public Understanding of Science 20(1): 48–63.

Besley

Nisbet

(2012) Predicting scientists’ participation in public life. Public Understanding of Science 22(8): 971–987.

Bucchi

(2002) Science in Society: An Introduction to Social Studies of Science. London: Routledge, pp. 70–82.

Bucchi

Saracino

(2012) Mapping variety in scientists’ attitudes towards the media and the public an exploratory study on Italian researchers. In: Bucchi

Trench

(eds) Quality, Honesty and Beauty in Science and Technology Communication. Vicenza: Edizioni Observa Science in Society, pp. 250–256.

10.

Burchell

(2015) Factors Affecting Public Engagement by Researchers: Literature Review. London: Policy Studies Institute.

11.

Burchell

Sheppard

Chambers

(2017) A ‘work in progress’? UK researchers and participation in public engagement. Research for All 1(1): 198–224.

12.

Crettaz von Roten

(2011) Gender differences in scientists’ public outreach and engagement activities. Science Communication 33(1): 52–75.

13.

Dubois-Shaik

Fusulier

(2015) Academic careers and gender inequality: Leaky pipeline and interrelated phenomena in seven European countries. Garcia Working paper n.5. Trento: University of Trento.

14.

Dudo

(2012) Toward a model of scientists’ public communication activity: The case of biomedical researchers. Science Communication 35(4): 476–501.

15.

European Commission (2019) She Figures Handbook 2018. Luxembourg: Publications Office of the European Union.

16.

Gibbons

(1999) Science’s new social contract with society. Nature 402: 81–84.

17.

Hamlyn

Shanahan

Lewis

O’Donoghue

Hanson

Burchell

(2015) Factors Affecting Public Engagement by Researchers. A Study on Behalf of a Consortium of UK Public Research Funders. London: TNS-BMRB.

18.

Howell

Singer

(2017) Pushy or a princess? Women experts and British broadcast new. Journalism Practice 11(9): 1062–1068.

19.

Jensen

(2011) A statistical picture of popularization activities and their evolutions in France. Public Understanding of Science 20(1): 26–36.

20.

Jensen

Rouquier

Kreimer

Croissant

(2008) Scientists who engage with society perform better academically. Science and Public Policy 35(7): 527–541.

21.

Johnson

Ecklund

Lincoln

(2014) Narratives of science outreach in elite contexts of academic science. Science Communication 36(1): 81–105.

22.

Kitzinger

Chimba

Williams

Haran

Boyce

(2008) Gender, Stereotypes and Expertise in the Press: How Newspapers Represent Female and Male Scientists. Cardiff: UK Resource Centre for Women in Science, Engineering and Technology and Cardiff University.

23.

Kreimer

Levin

Jensen

(2011) Popularization by Argentine researchers: The activities and motivations of CONICET scientists. Public Understanding of Science 20(1): 37–47.

24.

Kyvik

(2005) Popular science publishing and contributions to public discourse among university faculty. Science Communication 26(3): 288–311.

25.

Merton

(1986) The Matthew effect in science. Science 159(3810): 56–63.

26.

Mitchell

McKinnon

(2019) ‘Human’ or ‘objective’ faces of science? Gender stereotypes and the representation of scientists in the media. Public Understanding of Science 28(2): 177–190.

27.

Nichols

(2017) The Death of Expertise. The Campaign against Established Knowledge and Why It Matters. New York: Oxford University Press.

28.

Niemi

Pitkänen

(2017) Gendered use of experts in the media: Analysis of the gender gap in Finnish news journalism. Public Understanding of Science 26(3): 355–368.

29.

Peters

(2014) Scientists as public experts. Expectations and responsibilities. In: Bucchi

Trench

(eds) Routledge Handbook of Public Communication of Science and Technology, 2nd edn. London: Routledge, pp. 27–39.

30.

Peters

Brossard

de Cheveigné

Dunwoody

Kallfass

Miller

, et al. (2008) Interactions with the mass media. Science 321: 204–205.

31.

Poliakoff

Webb

(2007) What factors predict scientists’ intentions to participate in public engagement of science activities? Science Communication 29(2): 242–263.

32.

Rainie

Funk

Anderson

Page

(2015) How Scientists Engage the Public. Pew Research Center.

33.

Research Councils UK (2010) Concordat for engaging the public with research: a set of principles drawn up by the funders of research in the UK. https://re.ukri.org/documents/hefce-documents/concordat-for-engaging-the-public-with-research/

34.

Rossiter

(1993) The Matthew Matilda effect in science. Social Studies of Science 23(2): 325–341.

35.

Royal Society (1985) Public Understanding of Science. London: Royal Society.

36.

Scamuzzi

Tipaldo

(2015) Apriti scienza. Il presente e il futuro della comunicazione della scienza in Italia tra vincoli e nuove sfide [Open up science. The present and the future of science communication in Italy between constraints and new challenges]. Bologna: Il Mulino.

37.

Schoen

Theves

(Eds.) (2006) Strategic Management of University Research Activities. Methodological guide. Lugano: PRIME Project Observatory of the European University.

38.

Summ

Volpers

(2016) What’s science? Where’s science? Science journalism in German print media. Public Understanding of Science 25(7): 775–790.

39.

Thune

Reymert

Gulbrandsen

Aamodt

(2016) Universities and external engagement activities: Particular profiles for particular universities? Science and Public Policy 43(6): 774–786.

40.

Van Der Besselaar

Sandstrom

(2016) Gender differences in research performance and its impact on careers: A longitudinal case study. Scientometrics 106: 143–162.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB