Media Analysis on Perceptions of Worsening Social Order in Taiwan

Media Analysis on Perceptions of Worsening Social Order in Taiwan

What you need to know

Availability bias lies at the root of the feeling that there are more murders happening in Taiwan.

Each time the media reports on a horrifying murder, the public appears convinced that Taiwan’s social order is deteriorating. Conversations erupt on social media, particularly Facebook and Taiwan bulletin board PTT, decrying the collapse of society as we know it.

However, in reality, the number of violent crimes in Taiwan has been steadily decreasing, year after year. So what exactly is the reason for this perception-reality gap? The following analysis compares the actual number of crimes with the number of media reports on those crimes to investigate whether media coverage is behind the sense that safety in Taiwan has gone to hell in a hand basket.

Since the beginning of this year, there have already been 14 shocking cases of homicide, from the Hollywood-esque; encasement in concrete, to the grotesque; corpse dismemberment, which have resulted in the deaths of 19 people, with 14 others seriously injured.

This has resulted in a general feeling that there has been a surge of violent crimes and murders, and that Taiwan’s social order is crumbling. In addition, over the past few years, it seems that there have been horrendous murders every year, such as the high-profile Bali double homicide case, the Taipei metro rampage, the decapitation of a four-year-old girl, and so on, which have all lead to the elevated feeling that Taiwan’s social order is deteriorating.

However, judging from the National Police Agency’s (NPA) criminal case information, the number serious injury and murder crimes is at a 10-year low. From the data collected from the past 20 years, the number of violent crimes has decreased by more than 70 percent; from a high of 1,800 cases to only 432 last year.


The perception-reality gap and media reporting rates

Some people think that this is all down to the media, believing that in order to attract a greater number of readers, editors have chosen to issue an increasing number of sensational reports that have influenced society into thinking that murders are being committed at an exponential rate.

But let's look at the situation in terms of newspaper reports. Taiwan’s top three newspapers; Apple Daily, The China Times and the Liberty Times, have published all their newspaper headlines over the last several years online, so we can compile the data and calculate the frequency of reports on an annual basis.

The News Lens collected all of the headlines from all three newspapers and then used keywords related to serious injury or murder crimes to filter them. We then calculated how many days, within each year, the three newspapers reported on stories related to serious injury or murder crimes.

Before 2006, Apple Daily’s annual rate of reporting on violent crimes stood at more than 200 days per year, but the frequency began to gradually decline in the years following this.

Between 2007 and 2015, it fluctuated between 140-180 days per year, before declining significantly from 2016 onwards.


The Liberty Times had two instances where their annual rate was on an upward trend, the first being between 2006 and 2010, and the second from 2011 to 2016. In 2006, the number of days with related reports did not exceed 120 days, but after that, the annual rate of reports on violent crimes increased progressively, each year, until 2010, when it decreased slightly, before resuming an upward creep after 2011, averaging around 180-200 days per year.


The China Times had a gradual increase from 2010 to 2013, before witnessing a marked downturn, giving The China Times an average of about 130-180 days on which it covered reports of violent crimes.


We can see that between 2010 and 2015, you could read reports on violent crimes in both Apple Daily and The China Times for about a third of the year; and if you read the Liberty Times, then that proportion rose to half of the year.

From 2016 to 2017, the frequency of such reports decreased in each of the three newspapers. Apple Daily and The China Times both reduced their rates to between a third and a quarter of the year, while the Liberty Times reduced their rate to between a half and a third of the year.


So, let’s compare those findings with the number of actual crimes. Before 2010, the number of actual crimes was on the decline, and Apple Daily’s annual report rate tracked this trend, however, reports in the Liberty Times, instead, saw an increase.

Between 2010 and 2015, the total number of serious injury and murder cases plummeted from 768 to 464, at an almost consistent year-on-year rate, yet the coverage from all three newspapers remained pretty much in the same range as before.

However, since 2016, the frequency of sensational reports has greatly reduced at all three outlets.
Even though the annual report rate over the last two years has experienced a steady decline, there remains a sense that Taiwan’s social order is worsening. This could be related to the content of each report.

Influence of reported content: Availability bias

Recently, major crimes are being described in more gruesome and vivid detail, with the language used to describe them all leaning towards the sensational.

Such detailed descriptions and sensationalism only serve to strengthen our impression of the crimes, which reinforces our memory of them, and that in turn eventually leads to "availability bias" in our perception; that is, when our memories and impressions of something are deeper, our brains will mistakenly think that the chances of that thing happening are higher.

In his book, “Criminal: The Truth About Why People Do Bad Things,” author Tom Gash mentions that when people try to predict the probability of dramatic events happening, their efforts are often contrary to the facts; and especially when it comes to events related to violent crimes, people always seem to overestimate the incidence of these events. The report cited in the book as an example, reads: In 2010, 15 percent of Britons felt that they would be victims of violent crimes within the next year; but in reality, the probability was only 3 percent.

The book also mentions that the bias in thinking that leads to such overestimations, will also be influenced by fictional accounts. The year the movie "Jaws" was released, people were reluctant to go swimming at the beach, which lead to the number of swimmers on the California coast to plummet sharply, even though in reality there was no increase in the number of shark attacks.

In conclusion, it goes without saying that, though the number of violent crimes has decreased, they still continue to occur, and there will always be shocking events or actions that are incomprehensible to the general public. In the event of such a situation, instead of worrying about it, we should look to study and understand the social problems behind such events in order to provide better answers for addressing them in the future.

Read Next: TAIWAN: Despite Spate of Grisly Murders, Crime Data Makes for Reassuring Reading

The original (Chinese) version of this story can be viewed here.

Translator: Zeke Li

Editor: David Green