Heterogeneous response across genders to tonal variation in messaging: Experimental evidence

Vincenzo Galasso, Tommaso Nannicini

22 September 2013



Persuasion is an art which is critical to success in politics, business, and a personal career. ’Persuasive communication‘ – as defined by DellaVigna and Gentzkow (2010) – is used, for example, to convince:

  • Customers to purchase a new product.
  • A recruiting committee to award a promotion.
  • Citizens to vote for a candidate.

Most often this persuasion is exerted by individuals, firms or political parties who send competing messages to potential receivers.

A key decision in competitive persuasion is whether to be aggressive – running a campaign against potential rivals – or to be assertive – concentrating on self-promotion. For example, marketing strategies range from pure brand promotion to comparative advertising, political campaigns may feature both positive and negative advertising. Even coworkers competing for promotion may choose to praise themselves or to belittle others. 

  • To some extent persuasion styles are idiosyncratic. They reflect the characteristics of the individual party, or firm that sends the message.
  • The personal traits of receivers can be important in determining a persuasion campaign’s effectiveness; gender seems to be the most natural candidate.

For instance, should advertisers use the same arguments to convince female and male buyers, or should persuasive communication be gender specific?

A recent empirical literature indeed suggests that gender differences emerge both in the attitude toward competition and in the performance obtained in competitive environments (see Croson and Gneezy, 2009, and Bertrand, 2010).

Should we also expect men and women to react differently?

To tackle this question, we study gender differences in the response to competitive persuasion in a political campaign (Galasso and Nannicini 2013). We use a survey experiment in the field, run during the 2011 municipal election in Milan, to analyse the effect of positive vs. negative electoral campaigning on turnout and voting behavior of male and female voters. A seminal experiment by Ansolabehere et al. (1994) suggests that negative campaign messages reduce voter turnout, but does not study gender differences. Following studies have found mixed results.

Experimental design

We implemented our experiment by providing four surveys to an online sample of about 1,500 eligible voters. Respondents to the initial profiling survey – conducted at the end of March 2011 – were randomly assigned to two treatment groups and a control group exposed to neither campaign. There was one treatment group each for the positive and negative campaigns.

Individuals in the positive group were exposed to an electoral campaign with a positive tone by the main opponent, and those in the negative group to a campaign with a negative tone. Both treatment groups, as well as the control group, were exposed to the actual (non-randomised) campaign by the incumbent. The incumbent’s campaign was mainly perceived to be negative in tone by subjects in the control group.

Since actual political campaigns consist of various communication tools which potentially reinforce each other, we expose individuals in our sample to four devices of political persuasion:

  • A video interview with the candidate.
  • An electoral slogan.
  • An open letter to the voters; and
  • A video ad endorsed by the candidate.

Each of these items was presented to the two treatment groups in a positive or in a negative tone. Both positive and negative ads addressed the same issue, with the same format, and in the same setting (i.e. video images, length of the letter).

The initial two treatment tools (the video interview and campaign slogan) were provided with our second survey, run at the end of April 2011. The last two tools (the open letter and video ad) were provided with our third survey, run in the week before the election.

After administering each of the four tools, we used the corresponding survey to measure their instantaneous effect on the perceived credibility and approval rate of the candidates, as in a standard survey experiment. All campaign ads and videos can be watched on the experiment website.

The ’in the field‘ component of our experimental design comes from collecting (self-declared) turnout and voting choices through a fourth survey, run in the days immediately after the 15-16 May election. These responses enable us to evaluate the overall effect of our randomised campaigns on electoral behaviour.


Our empirical results show large differences in the gender response to political persuasion strategies. In fact, male and female voters respond in opposite ways to the degree of aggressiveness of the opponent’s campaign.

  • Negative advertising increases men’s turnout by about eight percentage points, but has no effect on women.

Gender differences are even stronger for electoral choices.

  • Women vote more for the opponent (by eight points) and less for the incumbent (by eight points) if exposed to the opponent’s positive campaign.

Exactly the opposite happens for men.

  • Men vote less for the opponent (by 11 points) and more for the incumbent (by 12.7 points) if exposed to the opponent’s positive campaign.

Overall, these effects amount to persuasion rates ranging from 21% to 24%.

Why do men and women respond differently to political advertisement?

Several channels may be at work.

  • Voters may cluster along gender-identity lines.

Since the 2011 election in Milan was a mixed-gender race between a female incumbent and a male opponent, women might have particularly disliked a negative campaign by a man against a woman. However, we exploit a 'natural' experiment, consisting of an attack by the (female) incumbent – Letizia Moratti – to her (male) opponent – Giuliano Pisapia during a campaign debate on Sky TV, to show that gender identification is not a first-order mechanism in our sample.

  • Women may have stronger preferences for redistribution and be more left leaning, therefore being more in favor of the (left-wing) opponent.

Although this correlation is confirmed in our sample, ideology only partially accounts for the higher responsiveness of women to the positive campaign run by the left-wing candidate.

  • Residual gender differences that might be interpreted according to the economic and neuro-psychological literature.

Women may just have a lower taste for competition  – especially if aggressive, both when they are active players – as shown by the existing literature – and when they are passive receivers of messages of competitive persuasion – as in our experiment (Paserman 2007).

Concluding remarks

The diffusion of social networks and the ability to process the huge amount of information collected in large datasets have allowed sellers and politicians to precisely identify their preferred targets: undecided, potential buyers, and swing voters. Our results suggest that since the ads can now be targeted to the right receiver, the message should be tailored to persuade her or him.


Ansolabehere, S, S Iyengar, A Simon, and N Valentino (1994), “Does Attack Advertising Demobilize the Electorate?” American Political Science Review, 88(4): 829–838.

Bertrand, M (2010), “New Perspectives on Gender”, in D Card and O Ashenfelter (eds), Handbook of Labor Economics, Vol 4, Part B, Amsterdam: Elsevier Ltd.

Croson, R and U Gneezy (2009), “Gender Differences in Preferences,” Journal of Economic Literature, 47(2): 448–474.

DellaVigna, S, and M Gentzkow (2010), “Persuasion: Empirical Evidence,” Annual Review of Economics, 2: 643–69.

Galasso, V and T Nannicini (2013), “Men Vote in Mars, Women Vote in Venus: A Survey Experiment in the Field” CEPR Discussion Paper No. 9547.

Paserman, M Daniele (2007), “Gender-linked performance differences in competitive environments: evidence from pro tennis”, VoxEU.org, 26 June.



Topics:  Gender

Tags:  gender, experimental economics, advertising, persuasive communication

Professor of Economics, Bocconi University; CEPR Research Fellow

Professor of Economics, Bocconi University