眾新聞 Logo
眾新聞 CitizenNews

Numbers Expose the Absurdity of Public Discussion in Hong Kong

By Four secondary school students

Liberal Studies has attracted much controversy over the years. However, both attacks against it and defenses of it have been mostly based on anecdotes or impressions.

In the following, we re-introduce an evidence-based approach to the public debate. Specifically, we use statistics compiled from the Liberal Studies public exam papers to examine three controversies that have long haunted the subject.

Liberal Studies Exam raw data and analysis.xlsx

The first controversy: Liberal Studies is more about opinions than facts

The first controversy we consider is that Liberal Studies favors the expression of squishy opinions over the analysis of hard facts. It rewards students for their viewpoints, but not for the cogency of their arguments or subject-matter knowledge. Professor Lau Chi-pang (劉智鵬), the chairperson of the committee responsible for reforming Liberal Studies, for example, has stated in a public seminar that the subject “prioritizes opinions and stances (意見行先、立場行先).” [1]

There are two reasons to reject this complaint. The first reason lies within the format of the Liberal Studies public exam. The marking scheme for Liberal Studies, published by the Hong Kong Examination and Assessment Authority, plainly states that students must use the sources given by the exam paper to answer questions. Personal feelings receive no marks.

For instance, the first question of the first Liberal Studies exam in 2012 asked students to “describe the trends of the population statistics of Hong Kong shown in Source A.” [2] Students cannot answer this question by resorting to their previous impressions of the city’s demographics. They have to analyze ‘Source A’ - a statistical table about Hong Kong’s changing population structure. Moreover, even for questions that require students to take a stance on a social issue, students still need to refer to the sources in the exam paper or face a marking penalty.

Although fact-based thinking, answering and marking are woven into the fabric of Liberal Studies and known to all students and teachers, it is still worth repeating this obvious point at the outset, given how misinformed some members of the public seem to be.

The second reason to view Liberal Studies as a fact-driven subject is that the composition of the given sources in the exam papers is heavily skewed towards hard facts. This has been common knowledge among Liberal Studies teachers but we are the first to quantify it here.

Categorizing the sources in the subject’s exam paper over the past 9 years, we found that around 70% of the sources were hard facts such as statistics, official reports, and news stories. About 30% were opinion sources, such as political cartoons or news commentary. Overall, hard facts outnumbered opinions two to one. (Graph 1)

Here, we would like to add, for those unfamiliar with assessment in education, that there is good reason for a significant minority of the sources to be subjective. Educational psychologists divide learning into a hierarchy of six learning levels: remembering, understanding, applying, analyzing, evaluating and creating. [3] 

Opinion sources allow students to be tested at the analysis and evaluation level. If students were only required to list or describe hard facts, there would be a performance ceiling, with most students clustering at a high score. Exam papers that test students with a mix of hard facts and opinions are more effective at distinguishing between students of different abilities.

The second controversy: Is Liberal Studies ‘too political’?

Another complaint is that Liberal Studies is ‘too political.’ In a recent opinion editorial in Ta Kung Pao, veteran Liberal Studies teacher Ng Pik-kin(吳壁堅)argued that too many exam questions were about Hong Kong politics. [4]  Dr. John Tan (陳岡),known as the ‘father of Liberal Studies,’ has also expressed a similar view. In 2014, he examined the Liberal Studies exam papers between 2012 and 2014 and found that the share of questions related to Hong Kong politics fluctuated between 20% to 37% - greatly exceeding the topic’s curriculum weight which is only 14%. [5]  More recently, in December last year, he drew on his 2014 analysis and expressed regret that the preponderance of political questions courted too much controversy and may have contributed to the subject’s demise. [6]

To examine this issue in greater detail, we extended Dr. Tan’s analysis for 6 more years to 2020. The full picture tells a different story. From 2015 onwards, the share of questions related to politics, though fluctuating from year to year, has generally shrunk. In fact, there were no questions about politics in both 2017 and 2018. (Graph 2)

Overall, between 2012 and 2020, political issues made up only 18% of the exam. For the five most recent years, it has been even lower. At an average of 14%, the relative size of political issues in the exam exactly mirrors official curriculum design.

Of note, we categorized exam questions rather promiscuously. Even when the question was only partly about politics, we labeled all the marks in the question as related to a political issue. For example, a question about economic competitiveness and democracy has both economic and political components, but we still labeled all marks of the question as related to politics. [7] The actual share of political issues is far below what we have presented above.

One could still argue that these findings mean little. The exams could be balanced; the classes need not be. Students might be taking a perfectly-balanced exam at the end of high school for a few hours, yet also sitting in a class dominated by political discussions for the preceding three hundred hours. Ultimately, this concern is a possibility, though one without any evidence beyond anecdotes. While exam data is not proof, it is still an important reference point that informs the larger debate.

The third controversy: Does Liberal Studies paint an unduly negative picture of China?

A third critique against Liberal Studies is that students are exposed to too much negative news about China. To name one example, Fanny Law (羅范椒芬), a member of the Executive Council, recently wrote a wide-ranging opinion piece in which, among other things, addressed this particular controversy. She examined ten Liberal Studies textbooks, and found that “some” contained “many” negative portrayals of China. [8]

Mrs Law’s willingness to use evidence is refreshing, but her argument would have been more persuasive if it were tabulated in numbers. It is not clear how much China-related content of a textbook has to be deemed negative for it to be judged as having “many” negative portrayals. The public is also left to guess the exact number of textbooks that fall into her negative-portrayal category.

To take a more constructive approach, we tallied all the Liberal Studies exam questions related to China and categorized them by emotional valence. The proportion of negatively charged questions would be a rough indicator of how China is portrayed in the subject.

We found that about one fourth of all China-related questions focused on negative aspects of China and less than 10% were positive. In relative terms, the weight of negative questions was close to triple that of positive. The sheer size of the discrepancy does indeed seem to bolster the view that China has been painted in an unduly negative light.

Nevertheless, a broader view of the data offers a less damning story. Negative questions and positive questions, in fact, only make up about one third of all China-related questions. The remaining two-thirds are questions that highlight both negative and positive aspects of China or those with no particular emotional valence. Looking at all four types of questions, only one-fourth are negative; three-fourths are not.

Another moot point is that most of the questions that have been categorized as negative are about policy problems that the Chinese government itself has set out to tackle, such as rural-urban disparity and the phenomenon of left-behind children. Claiming that students are tested on too many questions related to the development goals of their own country might be a little far-fetched.

How can we reshape public discussion?

In conclusion, we have responded to three common criticisms of Liberal Studies with preliminary statistics. Our data has cast doubt on the veracity of these complaints, to varying degrees in each case.

A more important point is that the Liberal Studies debate has been carrying on for years with almost no proper evidence. People who hold great sway over how hundreds of thousands of students are educated have justified their policy positions by declarative statements, vague adjectives and outdated statistics.

The late Harvard statistician Frederick Mosteller famously said that “it is easy to lie with statistics, but easier to lie without them.” This truism highlights that, when passions run high, numbers - in its disciplined and precise language - serve as imperfect guardians against our most outrageous presumptions. At its best, numbers can even be a platform for earnest debate and civic engagement. Paying heed to such wisdom perhaps would allay some of the divisiveness that has been afflicting Hong Kong. 


1: “Teachers claim that it would be difficult to develop critical thinking in the new Liberal Studies; Liberal Studies reform committee: old subject prioritizes stances,” Ming Pao, 2021-2-11. 〈教師指新通識難練思考 冠名會:舊科立場行先〉,《明報》

2: Question 1(a), Paper 1, 2012 Liberal Studies DSE Exam

3: Bloom, Benjamin Samuel, David R. Krathwohl, and Bertram B. Masia. 1986. Taxonomy of educational objectives : the classification of educational goals. New York: Longman.

4: Ng Pik-kin, “We need to resolve the problem of the Liberal Studies exam being too heavily skewed towards political issues,” Ta Kung Pao, 2019-11-1. 吳壁堅,〈要解決通識科試題偏重政治的弊病〉,《大公報》

5: John Kang Tan, “Protests are the fruits of learning Liberal Studies? Exploring the deep-rooted reasons that explain the subject’s bias towards political issues,” Ming Pao, 2014-9-25. 陳岡,〈抗爭是學習通識的成果?——探討「通識科偏重政治」的深層原因〉,《明報》

6: John Kang Tan, “Reforming Liberal Studies could have been initiated by the Hong Kong Examination Authority: replying to the Chief Executive’s comment that Liberal Studies has devolved,” Ming Pao, 2020-12-8. 陳岡,〈改革通識科 本可由改革考局教局人事起——回應特首提出「糾正通識科異化」〉《明報》

7: Question 2(b), Paper 1, 2016, Liberal Studies DSE Exam. “Source B shows the ranking of the Democracy Index and that of the Global Competitiveness Index. Do you think that there are any relationships between the two rankings? Explain your answer using Source B.”

8: Fanny Law, “Liberal Studies: to reach our goal, be unswerving,” Ming Pao, 2020-12-4.  羅范椒芬,〈談通識教育科:不忘初心,方得始終〉,《明報》

The original Chinese version of the article 〈4名中學生:被拋棄的數據——從通識教育的三大爭辯看公眾討論的荒誕〉 was published in Ming Pao on 7th April 2021.

請加入成為眾新聞的月費訂戶,長期支持我們的工作。所有訂戶都可以收到我們的「每周時事」通訊 。