• Citer

Essays on Communication Effectiveness in Non-profit Organizations

par Esterzon, Emilie
Président du jury Bruneau, Virginie
Promoteur Rothenberger, Sandra
Publication Non publié, 2022-04-04

Thèse de doctorat

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

As it is for any institution, communication is very important for non-profit organizations. In this thesis, I study the effects of communication on behavior. More specifically, I study which elements of a communication drive individuals into the desired behavior. I rely on literature on consumer behavior and on experimentation methods to shed empirical light on the effects of communication on behavior. I first look at the impact of text messages on attendance to medical appointments. The database contains about 600.000 medical appointments made by more than 40.000 individuals during a period of 12 years. At a certain point in time, the hospital launched text message reminders to increase attendance. The analysis empirically confirms that reminders have a positive effect on attendance rate and reveals three noteworthy observations: The effect is stronger among young people, people living near the hospital (as if the geographical proximity used to reduce the importance given to the appointment while the geographical distance increased it) and in less vital medical services, such as dermatology. I then focus on fundraising campaigns for non-profit organizations. Although non-profits are increasingly turning to digital channels, written communication (direct mailing) remains the primary fundraising channel for many of them. In a randomized control trial among more than 40.000 prospective donors of an international NGO based in Brussels, we manipulate the opportunity to target a gift by providing a physical choice between causes and by associating each cause with a suggested donation amount. We find that targeting increases the propensity to donate and the amount donated: The interventions nudge donors to contribute the donation amounts suggested by the charity and boost total fundraising revenues by 42%. Yet, using machine learning for causal inference, we find substantial heterogeneity in the effect of both interventions, highlighting potential negative effects of the targeting interventions. More experienced donors did not respond positively to the suggested amounts; relatively inactive donors did not appreciate to be forced to choose a cause; and less generous donors rejected both targeting interventions. I then analyze 3 years of direct mailing campaigns by 4 different NGOs. Using text mining algorithms, we identify attributes of direct mailings that influence fundraising effectiveness, regardless of the characteristics of the individuals who received the communication. We then estimate econometrically the impact of these attributes on campaign effectiveness. Overall, the thesis shows the importance for non-profit organizations to target the right person, with the right content, at the right moment. Conclusions provide guidance to non-profit organizations on how to improve their communication effectiveness. The thesis also opens avenues for further research on the determinants of heterogenous responses to specific elements of a communication.