The Entertaining Way to Behavioral Change: Fighting HIV with MTV

"The enormous potential reach in terms of number of viewers and the low marginal costs of distribution make edutainment communication tools potentially very valuable for development policy."

This paper shares the results of a randomised controlled trial (RCT) testing the effectiveness of the third season of the entertainment-education (edutainment) television series MTV Shuga, which sought to provide information and change attitudes and behaviours related to HIV/AIDS. (See Related Summaries, below.) Following an introduction, section 2 of the paper presents a stylised model that guides the empirical analysis; section 3 describes the experimental design; section 4 details the empirical strategy and data; sections 5 and 6 present results on the basic treatment and on social effects, respectively; section 7 contains robustness analysis; and section 8 concludes the paper.

As the researchers explain, there are two reasons why edutainment might work where ordinary behaviour change campaigns fail. First, the appeal of the show could make the individual pay more attention to the message and reduces potential resistance to top-down advice (the "individual effect"). Second, the "social effect" comes in because, if people conform to what others do or think, the message on the screen, potentially coupled with the fact that television shows get seen by a large number of people, could coordinate a shift in the social norm. However, whether edutainment actually works or not in these ways is an open question.

Produced by MTV Staying Alive Foundation, MTV Shuga presents young Africans from various socioeconomic strata whose bright futures could be compromised by the negative consequences of high-risk behaviours. Filmed in Nigeria in 2013, Shuga's third season features prominent Nigerian actors and music. In order to have exogenous variation in the exposure to the show, the researchers conducted the study before MTV Shuga was widely distributed in Nigeria, organising screenings in community centres, schools, and other locations.

The RCT covered over 5,000 young men and women in 80 urban and peri-urban locations in South West Nigeria: 54 locations were randomly assigned to screen MTV Shuga, while the remaining 26 screened a different serial that involved a similar demographic but made no connections to HIV. In short:

• Treatment 1 (T1) consisted only of MTV Shuga screenings and was administered in 27 randomly selected locations.
• T2 involved another 27 randomly chosen locations and was the same as T1, except that after the MTV Shuga episodes, researchers showed video-clips including interviews of youth condemning negative behaviours and praising positive ones after watching Shuga, as well as statistics, to show how common certain beliefs and attitudes are among other participants and how willing they may be to change them.
• T3 involved randomly selecting half the treated individuals and offering them the option of bringing up to two friends to the screenings. The goal of this treatment was to determine whether the effect of Shuga differs when individuals can discuss its content with close peers who also watched the drama.

Thus, participants in the experiment remained "private" in the sense that their choices were not announced to others: the concern that their behaviour may be observed and sanctioned by others may emerge in the long run but was not a direct consequence of the experimental design. What the experiment randomly makes public is information on the beliefs and behaviours of others.

The researchers found "striking effects on knowledge about sources of transmission of HIV and its treatment, on attitudes towards HIV+ people and on a range of behavioral outcomes (both self-reported and objectively measured)" 8 months or more after the showing. For example, the likelihood of testing for HIV, objectively measured through redemption of a voucher that the researchers distributed at health camps, increased by 3:1 percentage points in the treatment compared to the control group. This corresponds to a 100% increase over the control group mean; so, treated subjects are twice as likely to get tested for HIV. (Analogous effects are estimated for the self-reported measure, where the likelihood of testing increases by 2:5 percentage points.) Corresponding to this effect is an improvement in treated individuals' knowledge about HIV, including sources of transmission, awareness of anti-retroviral drugs, and the need to take a second HIV test after at least 3 months from the first (the "window period"). These are topics specifically covered in MTV Shuga. The effects "are robust to aggregating outcomes into indexes and to correcting for multiple hypothesis testing."

The researchers found more nuanced effects on risky sexual behaviour. While the acceptability and reported incidence of concurrent sexual partnerships significantly decreased, MTV Shuga did not induce greater condom use - neither as reported by respondents nor as revealed in an experimental game that the subjects played in health camps. Yet, there were significant impacts on a biomarker that proxies for unprotected sex with risky partners, and the likelihood of testing positive for the sexually transmitted infection (STI) Chlamydia decreased by 55% in response to treatment for women in the sample (the impact on men is in the same direction but statistically insignificant).

The endline survey included questions about how immersed in the story the respondent was while watching the show, and how much he or she identified with the characters. The researchers then constructed two indexes, "Transportation" and "Identification". The treatment effects were stronger for viewers that had higher values of these indexes, which is consistent with the psychological underpinnings of edutainment. "The results support the view that edutainment needs to be absorbing in order to work."

The above findings shed light on the proposed individual effect of edutainment, which is shown to have been strong. The researchers also looked into the role of social effects. (The views and behaviours portrayed in MTV Shuga could signal a new norm, such as about how one should interact with HIV-positive people and with sexual partners (inductive norm). In addition, viewers may have expected Shuga to have wide viewership, and they may have internalised the shift in norms that would take place as a result (descriptive norm). To examine this possibility, they took several approaches, which are detailed in the paper. In short, the trial's experimental manipulations of the social norm component did not produce significantly different results from the main treatment. This "suggests that coordination on social norms was not a big part of the effect of MTV Shuga. This could be because viewers do not care about social norms when it comes to very private decisions, like ones relating to HIV and risky sexual behavior."

Takeaway messages include:

• The individual effect of edutainment prevailed in the context of this study, which suggests that - at least in the context of HIV/AIDS - people react to the messages they see on TV regardless of what others say.
• If policymakers wanted to leverage social effects in edutainment, they should experiment with different (and potentially larger) "reference groups" (a collection of people we use as a standard of comparison for ourselves regardless of whether we are part of that group) - e.g., schools, villages, etc.
• "[M]ore research is needed to assess the potential role of conformity when the private versus public nature of the message is varied. This seems especially relevant for the edutainment agenda given the growing importance of social networks in today's society."

Editor's note: Please see these related, additional documents/resources:

• The Entertaining Way to Behavioral Change: Fighting HIV with MTV [PDF], by the Narrating Behavior Change programme of DIME, 2019;
• The entertaining way to behavioural change: Fighting HIV with MTV in Nigeria, by Abhijit Banerjee, Eliana La Ferrara, and Victor H. Orozco-Olvera, VoxDev, May 23 2018;
• Watch The MTV Soap Opera That Is Secretly Teaching Sex Ed, by Maanvi Singh, NPR, April 18 2017;
• MTV Shuga Soap Opera Turns Edutainment into a Tool to Fight HIV and Gender-Based Violence, WB website, February 23 2017; and
• Podcast (2017), in which Banerjee and Orozco explore edutainment's potential to motivate people to change their behaviour and to tackle development problems such as HIV/AIDS, in the context of the above-summarised study.