A full report and set of interactives will be released in May 2023
Work conducted by Vrinda Bhat, Pooja Dantewadia, Michael Krisch, Katharina Tittel and Katie Watson
The Brown Institute, in partnership with the United Nations Population Fund, is excited to announce a research endeavor investigating and visualizing the patterns and trends of media coverage on population and fertility across eight countries, including the United States, Brazil, France, Hungary, Japan, Nigeria, India, and Egypt; countries identified by the UNFPA for a public perception survey conducted for the flagship 2023 State of World Population report. Drawing on high-circulation news organizations as sources, specifically those identified in the Reuters Digital News Report, this study examines over 11 million news stories from 2020 through 2022. At its core, it looks to uncover the prevalence of population and fertility discussions within the larger media discourse in the different countries. Subsequently, when discussing population or fertility rates, what is the framing and tone of coverage? For example, the study examines the co-occurrence of population and economic concerns, fertility and health and education access, and population and migration, analyzed across countries. It also seeks to explore the article’s sentiment, highlighting what media coverage looks like at a country level. Through advanced data analysis and visualization techniques, the findings of this project will shed light on the patterns and dynamics of media coverage of population-related issues across diverse cultural and geographic contexts, providing valuable insights into how these issues are framed and discussed in the public sphere. Ultimately, this research will contribute to a more nuanced understanding of the intersections between media and public perception on topics of population and fertility.
As the world population surpasses 8 billion, it is crucial to understand how fertility and population are reported on both at the local and global level. Commissioned by the United Nations Population Fund (UNFPA), this research delves into the media’s portrayal of population growth and its implications. When launching this work, the focus was on news coverage surrounding November 15, 2022, the date on which the world population reached this historic milestone. Through our initial analysis, we recognize the significance of taking a more longitudinal approach to studying reporting patterns, which will enable us to contextualize this event within larger trends and explore other important factors shaping the conversation around population growth. In this report, we present our initial findings on the representation of fertility and population in the media, as well as provide insights that might be used to improve discourse and create more informed dialogue around these critical issues. A more detailed report will be released in May, expanding on the analysis found below.
To investigate how perceptions of fertility and population were reported in the media, we analyzed a vast dataset of news articles sourced from GDELT, an online news database. The GDELT Project, supported by Google Jigsaw, monitors the world’s news media in print, broadcast, and web formats, in over 100 languages, and collects and publishes content on a 15-minute cadence. Our research drew on a survey of 11 million articles published between January 1, 2020 and December 31, 2022, encompassing total coverage from 63 of the highest-reach news outlets, including both online and print platforms. The countries we focused on were the United States, Brazil, France, Hungary, Japan, Nigeria, India, and Egypt. For each of these countries, we selected a combination of the top 5 online outlets and top 5 TV, radio, and print outlets, with the exception of Egypt, which required additional consultation with a media researcher. News organizations topping the Digital News Report lists in multiple countries and shared domain prefixes such as Globo account for the discrepancy and final count of organizations being analyzed.
In total, 491,224 news stories related to the themes of population and fertility were identified using extensive keyword match techniques performed by the GDELT service, which translated article text into English for analysis. In order to ensure that our analysis focused on the most relevant and accurate news stories, we employed a sampling process to identify topics most closely related to our themes. Specifically, we analyzed an initial set of 10 related topics and identified a subset of four with a relevancy rate of 95% or greater across a 200-article sample. In addition to theme identification, GDELT assigns emotional dimensions to each article, identified on the original language text. For the purpose of our analysis, we made use of tone and polarity measures described later in the report.
The impact of the Covid-19 pandemic on the counts of news stories related to population and fertility is striking. When looking across samples of headlines, there are few reasons to exclude pandemic-related reporting, as most made explicit reference to the national or global population, the impacts of the virus and vaccine on maternal and child health, and on death rates and lockdowns and a collapse in fertility rates in countries around the world. Trends in relevant reporting, shown below, help expose the relationship between the Covid-19 pandemic coverage and our theme.
We conducted a range of analyses on our extensive dataset. Our research encompassed a country-by-country analysis of relevant articles, an exploration of the frequency in which external countries were referenced in articles pertaining to population and fertility, an examination of various subthemes covered when reporting on the overarching theme, and a series of case studies including the intersection of population and fertility coverage in US and Brazil elections. By examining a diverse subset of analysis, the research provides a detailed understanding of how the media has approached and framed discussions surrounding population growth and fertility, both globally and within specific countries and contexts.
Coverage and Relevance
Relevant coverage to the theme of population and fertility largely followed patterns of coverage relating to the Covid-19 pandemic. As referenced above, there is little reason to exclude this reporting, as it directly references national and global populations and required national and global public health interventions. Each of these interventions had specific and far-reaching impacts, ranging from regressions in life expectancy to continued and dramatic drops in fertility rates in many countries.
In the early months of 2020, coverage relating the theme was low and stable but saw a marked uptick at the start of the Covid-19 lockdowns, happening in January 2020 in Wuhan and in mid-March in the countries included in this study. As infections and deaths compounded worldwide, so too did news coverage related to population. In addition to these indicators, vaccine development and rollout was also a significant driver of coverage. When looking at country-specific trends, we observed spikes in coverage occurring alongside spikes in new infections. For example, coverage in Brazil followed surges in infections within the country. Another example can be seen when analyzing India and its vaccine rollout and surges in new infections due to the Delta and Omicron variants. Relevant population and fertility coverage, as well as new the Covid-19 pandemic infections in India, can be seen in the two charts below.
These findings underscore the significant impact the global pandemic had on media related to population and demographic-related, and how difficult it is to parse the two. It also highlights how non-Covid-19 pandemic coverage related to more niche reporting on fertility or country-based population measures was overshadowed during this time.
A detailed table of theme-relevant coverage can be found in the appendix. A summary of coverage can be seen below.
|Country||Coverage Counts||Coverage Share|
For sentiment analysis, we evaluated the mean tone of articles as seen at a country, theme, and subtheme level. The tone metric is represented as the average sentiment of the document as a whole, ranging from -100 (extremely negative) to +100 (extremely positive), with common values falling between -10 and +10, with 0 indicating neutrality. It is important to note that documents with a tone score close to zero may have a low emotional response or roughly equivalent positive and negative scores, nullifying one another.
Analyzing coverage en masse helps to highlight the variation in tone across news organizations and countries. To showcase this variation across countries, we calculated the mean tone of all articles related to the theme (~491K articles) and compared it to the mean tone of unrelated articles (~10.5M articles). The differences between the two are stark.
We are actively working to produce more in-depth tone analysis in the upcoming weeks, which will include exploring variations in tone across subthemes. Anecdotally, and in analyzing a subset of Egyptian media, migration-related coverage tends to be most negative across all articles, and economic, environmental, and educational stories tend to be more negative when discussing population. Validation and examples supporting these findings is underway.
In linking this investigation with the UNFPA public opinion survey, co-occurrence analysis of subthemes related to population and fertility within the extensive dataset of news articles is critical. Specifically, we examined the frequency of seven distinct subthemes: food, the economy, conflict, the environment, urbanism and housing, education, and migration. These subthemes were selected in alignment with the YouGov public perception survey conducted by the UNFPA, which aimed to identify areas of hope and concern in the context of population and fertility.
Our analysis revealed several key takeaways regarding the co-occurrence of subthemes related to population and fertility within news coverage. Perhaps unsurprisingly, the most commonly referenced subtheme across all countries and regions was the economy, which appeared in 44% of articles analyzed. Notably, Japan and Nigeria had an even higher percentage of stories referencing the economy, at 49% and 48%, respectively, suggesting that discussions of population and fertility are often framed within economic contexts in these countries.
In addition to the economy, we found that the environment was also a common subtheme, and in Brazil, Hungary, and Egypt, it was the most prevalent at 51%, 45%, and 56%. These suggest that concerns around environmental sustainability is a substantial concern within these countries in the context of population.
We also spent significant time exploring the occurrence of subthemes across our two corpora of data: the ~491k news stories related to population and fertility and the ~10.5M news stories not related to this theme. Preliminary analysis suggests a significant difference in the patterns of subthemes across these two sets of articles. For instance, we observed substantially higher levels of education-related content in population and fertility coverage than in non-relevant coverage (14% difference, on average). Additionally, both the US and Japan exhibited higher frequencies of reporting on the subthemes of conflict (10% and 6%, respectively) and migration (7% and 11%) in population coverage relative to unrelated coverage. On the other hand, France and Nigeria displayed lower levels of co-occurrence between conflict and population coverage (-3% and -6%). Each of these findings are intriguing, and we plan to produce more in-depth analysis in the coming weeks.
Additional research that is currently underway will explore the role of images in news coverage related to population and fertility. In this work, we will conduct a visual analysis of headline images and the corresponding news headlines to identify patterns and themes that may not be evident in text alone. In addition, we are producing case studies on population and fertility coverage around major events, including national elections, protests, and other political events. Lastly, we are exploring references to other external countries in coverage, specifically in the context of subthemes. Publication of this work is expected in early May 2023.
|country||site||Population articles||Total articles||Share population-related|