Non-Governmental Organizations (NGOs) play a crucial role in disseminating information about children's rights and mobilizing public attention to their plight. This is particularly important given the vulnerability of children globally and the limitations of public policy alone in addressing their needs. The Sustainable Development Goals (SDGs) and the Convention on the Rights of the Child (CRC) highlight the international commitment to protecting children's rights, but NGOs are essential partners in this effort. Social media, particularly Twitter, offers a valuable tool for NGOs to communicate directly with the public, engage in dialogue, and raise awareness. While previous research has examined NGO communication strategies, there's a lack of understanding on how Spanish child protection NGOs leverage social marketing principles and prospect theory on Twitter to maximize their impact. This study aims to fill this gap by analyzing the communication profiles of leading Spanish NGOs on Twitter, classifying their messages, assessing message effectiveness, and examining organizational positioning based on message type.

The literature review examines the roles of NGOs in child protection, social marketing principles and their application in changing behaviors, and prospect theory's relevance to understanding decision-making in the context of risk and uncertainty. It discusses the challenges NGOs face in effectively using social media for communication, highlighting the importance of strategic planning and understanding the nuances of different social media platforms. The review also explores the existing research on the use of Twitter by NGOs and the limited understanding of how message characteristics (informativeness, entertainment, credibility) affect user engagement. The role of happiness in social marketing and its impact on user behavior and loyalty are also addressed, emphasizing the scarcity of research connecting happiness, social marketing, and NGO communication.

This research employed a descriptive and correlational approach using data mining (Knowledge Discovery in Databases - KDD model), content analysis, and simple correspondence analysis (SCA). Data was collected using Fanpage Karma, extracting tweets from November 19, 2022, to March 31, 2023, from seven prominent Spanish child protection NGOs (Ayuda en Acción, Educo, Fundación Tierra de Hombres, Plan International España, Save the Children España, UNICEF España, and World Vision España). The initial 4,213 tweets were filtered to exclude retweets, and a representative sample of 1,035 tweets was analyzed. Content analysis involved coding tweets based on format, purpose (informative, dialogue, behavioral), topic, sentiment, alignment with prospect theory, and number of likes. Two SCAs were conducted: one to classify message types based on purpose and prospect theory alignment, and another to analyze NGO positioning based on message type. The Kruskal-Wallis test was used to assess the relationship between message type and the number of likes.

The analysis revealed that NGOs generally receive a low number of likes relative to the number of tweets published, except for Save the Children España. Emergency situations (e.g., earthquakes in Turkey and Syria) resulted in a significant increase in both tweets and likes. About 43.3% of tweets contained emotional connotations (27.45% positive, 15.85% negative), with impartial messages receiving the highest average number of likes. Informative tweets comprised 72.4% of the total, predominantly from UNICEF España and Ayuda en Acción, yet behavioral tweets had a greater impact. Over half of the tweets (55.1%) were impartial regarding prospect theory, not highlighting risks or benefits. Tweets highlighting risks were more frequent than those emphasizing benefits, aligning with prospect theory's risk aversion premise. Messages about collaboration, natural disasters, and gender equality received the most likes. SCA identified three message types: Type 1 (behavioral, benefit-focused), Type 2 (dialogue, impartial), and Type 3 (informative, risk-focused). Type 3 messages had the highest average number of likes, followed by Type 1 and then Type 2. The final SCA illustrated NGO positioning relative to these message types, showing some organizational preferences (Educo – Type 1, World Vision Spain – Type 3, UNICEF Spain – Type 2), but no clear clustering.

The findings indicate that while NGOs utilize Twitter extensively, their communication strategies could be optimized to increase engagement. The higher impact of risk-focused and benefit-focused messages aligns with prospect theory, suggesting NGOs should tailor their messaging to leverage this effect. The low impact of impartial dialogue tweets highlights the need for more interactive and engaging approaches. The success of messages during emergency situations underscores the potential of social media for rapid response and resource mobilization. The study contributes to a better understanding of NGO communication practices and offers practical guidelines for improving the effectiveness of social marketing campaigns on Twitter.

This research demonstrates the value of utilizing social media analytics to understand and improve NGO communication strategies. The identification of three distinct message types and their varying impact highlights the need for tailored approaches to maximize engagement. The findings suggest focusing on behavioral messages that highlight either risks or benefits while minimizing impartial dialogue. Future research could explore the impact of various message characteristics across different social media platforms and demographics, including age and cultural differences.

The study is limited to a specific timeframe and sample of Spanish NGOs and their Twitter activity. The generalizability of the findings may be limited to similar contexts. A larger, more diverse sample of NGOs and social media platforms would strengthen the study's conclusions. Further exploration into the long-term effects of different message types and the influence of external factors on user engagement is warranted.