Studying the Science of Happiness

Associate Professor Christie Scollon at the SMU School of Social Sciences has been studying cultural differences in subjective wellbeing for more than a decade. In particular, she is interested in understanding the cultural differences that cause Asian cultures to report lower levels of happiness than other places.

Photo Credit: Darren Yau

By Dora Yip

SMU Office of Research (30 Sep 2013) – The quantification of happiness – few problems could have a greater contrast between their global importance and ease of reduction to statistical rigour. Is it even possible to distill our ‘happiness quotient’ and study large groups of people in real time, to help policy makers and researchers gain insight into people’s lives?

Those are the goals of Associate Professor Christie Napa Scollon at the Singapore Management University’s School of Social Sciences, who is an expert on the topic of subjective wellbeing, or happiness.

“I have been interested in cultural differences in subjective wellbeing for over a decade now. There have been several studies showing cultural differences in emotions and happiness, and my work builds on this by trying to understand the causes of the cultural differences,” she says.

Findings from several studies have shown national differences in emotion and life satisfaction and Professor Scollon is especially keen to understand why Asian cultures report lower levels of happiness than other places.

“Even within a single country like the United States, Asian Americans report lower happiness than Whites or other minorities. As an Asian American, I found this really intriguing.”

While some of the national differences in happiness can be explained by differences in Gross Domestic Product (GDP), what stood out to Professor Scollon was that countries with a high GDP like Japan, Taiwan, Singapore and South Korea had a surprisingly low level of life satisfaction when compared to countries with a lower GDP like those in South America.

“It’s not that the Asian tigers are the most miserable in the world. But what I found intriguing is that life is not as happy in these places as you would expect based on GDP or based on individualism-collectivism. So my work has sought to explain the surprisingly low subjective wellbeing in these places. I haven’t come up with all the answers, but it seems that part of these differences has to do with measurement and part has to do with values and norms. In Asia, positive emotions are not valued as much as in North America,” she explains.

“Emotions are also a much more socially embedded experience in Asia than in the US,” she says. In one such study, Professor Scollon and her PhD student Sharon Koh measured the cognitive organisation of emotion knowledge using a reaction time paradigm (originally developed by Dr Michael Robinson at North Dakota State University). By asking participants a series of questions and measuring how quickly they responded, they made some inferences about how Singaporeans and Americans organise their emotions.

“We used reaction times to understand how tightly networked ideas and connections are. In Singapore, people organise emotions around people. So it’s not just ‘I’m a happy person’, but ‘I’m happy when I’m with my friend Jane, for example’. It is relationship dependent. In America, emotions are less organised around relationships, and more organised around the person. So it’s more a case of ‘I am a happy person, it doesn’t matter who I’m with’.”

Making sense out of a haze of data

In addition to her studies on subjective wellbeing, Professor Scollon collaborated in June this year with the Living Analytics Research Centre (LARC) to study how Singaporeans responded to the severe haze that shrouded the country. The LARC is an interdisciplinary team of a dozen researchers, including computer scientists and engineers, that is headed by Professor Lim Ee-Peng from the SMU School of Information Systems.  The LARC team analysed data generated by about 130,000 Twitter public users in Singapore to derive real-time insights about people’s reactions to the haze.

The researchers began by looking at how the haze affected Twitter usage in general, and found that overall daily tweets surged on 17, 19 and 21 June when the Pollutant Standards Index (PSI) hit record numbers at 111, 321 and 401 respectively.

Next, they examined a subset of tweets that mentioned key haze-related words and hash tags to determine if the increased tweeting was due to the haze. Again, the pattern of haze-related tweeting activity mirrored the PSI score. About 45,000 and 58,000 users generated more than 133,000 and 204,000 haze-related tweets (8 to 10 percent of all tweets) on 17 and 19 June respectively. After 21 June when the haze began to subside, the number of haze-related tweets decreased significantly, but remained higher than before the haze.

Finally, the team examined all tweets during the haze period for emotion words, which were categorised into anxiety-related words (e.g. “worried”); anger-related words (e.g. “irritated”); and swear words/exclamation (e.g. “damn”, “oh no!”). They found that negative emotions, especially swear words and exclamations, increased as the haze continued and worsened, peaking on 19 June, the day the PSI hit record highs.

“People were very negative about the haze, very angry and upset,” she says. “It also restricted their physical movements, which gives us hard data on the economic fallout.”

Of note is that the Twitter research allowed researchers a window into the thinking of a large group of people in real time. For the haze study, the sample group was far larger than most large-scale surveys. But beyond studying the impact of the haze on Singaporeans’ lives, studies like these may have the potential to help policy makers and consumer researchers.

“It's always a bit hard to predict how certain technology could be used in the future – after all, I don't think when Twitter was invented anyone thought it would be used for research,” she says. “If there is a specific event, say a natural disaster or an election, I think the technology could be useful for policy makers who want to gauge public reactions.”

That said, Professor Scollon adds that more work needs to be done before we can extrapolate data from Twitter into predictive models, or study riots before they happen. She does, however, see it being used in consumer research, allowing researchers to capture people’s emotions shortly after they check-in to a movie theatre or restaurant for example.

With advances in cognitive research, Professor Scollon is confident that the measurement of emotions and wellbeing will continue to improve. But the challenge remains, though, of communicating the findings to non-scientists and policy makers in an accessible and meaningful way. After all, subjective wellbeing has implications for everyday lives and society at large.

“Who doesn’t want to be happier?” she asks.

Listen to podcast