Yawning Bread. August 2007

Singapore second last in Asia democracy index




I was a well-wisher at the 27th anniversary dinner of the Singapore Democratic Party last Saturday night, and in return, I got a book, Asia Democracy Index 2005. I found it so engrossing that I started reading it even as speeches were being made. Yes, it was a little rude of me.

The summary index showed Singapore scoring second from bottom among 16 countries/territories. We were outdone only by Burma (Myanmar).

Within the book's 300-plus pages was a treasure trove of data from a survey conducted in the first half of 2005. They had wanted to do more than these 16 countries, but budget, logistical or safety issues prevented them from covering some places.

"They" and "them", I think, need to be carefully spelt out. The book and the survey were commissioned by the Alliance for Reform and Democracy in Asia (ARDA), based in Singapore and chaired by Chee Soon Juan, who is also the Secretary-General of the opposition Singapore Democratic Party.

However, the survey design team comprised 11 human rights activists, lawyers and academics, only one of whom is known to be Singaporean or based in Singapore. The team included Law Yuk Kai, the director of the Hong Kong Human Rights Monitor, Paul Scott, a professor at Kansai Gaidai University (Japan), and Nadine Kreisberger, an advisor to the Vice-Speaker of the Mongolian Parliament. Ms Salbiah Ahmad, a lawyer, is the Singaporean in the team, though she's based in Malaysia and better known as the founder of the Malaysian NGO, Sisters in Islam.

Likewise, from a cursory look at their names, the 7-member editorial board for the book did not include anyone from the Singapore Democratic Party.

Survey method

As with all surveys conducted across such different countries, from Nepal to Japan, it was hard to maintain a strictly uniform survey method, especially when budget seemed to be tight. Basically, however, it involved a partner organisation or individual researcher in each country interviewing a minimum of 100 respondents, using a standardised questionnaire with 22 questions.

How those respondents were selected seemed to be so variable that one has to be careful in interpreting the results. The only requirement set out seemed to be to poll "politically-aware" people in the country. How to define "politically aware" or to find them was left to each country researcher.

In Bangladesh, the partner organisation interviewed 110 "local-level leaders of major political parties, teachers. small businessmen, journalists, civil society participants and students."

In Indonesia, they "included reporters, student activists, pro-democracy and other social movement activists, political party activists and/or cadres, as well as professionals, business people, etc."

In some countries, the survey-takers around the cities and provinces were required to ask pre-qualifying questions. In Myanmar, "respondents had to answer correctly to ... three questions and listen to radios more than once a week to be considered politically aware", the 3 pre-qualifying questions being: "identifying the status of a politician"; knowing the name of the country's ruling committee; and "awareness of a recent political event".

Likewise in Thailand, the 2 pre-qualifying questions were: "Name at least 3 out of 4 political parties" that were then represented in Parliament (the survey was done during the Thaksin administration. before the September 2006 coup); and "State the name of the leader of the Opposition parties". The survey was conducted by graduate students from the Faculty of Political Science, Chulalongkorn University, fanning out to several provinces, under the leadership of an assistant professor and a lecturer.

The chapter on Singapore gave an unacceptably brief description of the sample selection process:

The survey was conducted by the Association for Democracy in Singapore during the months of April and May 2005. Taking part in the survey were 148 people, comprising members of the general public, university students, NGO members, lawyers and journalists. They were solicited at public forums, university campuses and private meetings.

Somewhat amazingly, far-off Mongolia did best. Its survey description, 5 times longer than Singapore's, said that it was carried out by a team of statisticians and political scientists from the National University of Mongolia, with an initial total of 802 respondents. They used 4 introductory questions to screen the politically aware from the unaware, reducing the 802 to 556 respondents.

That said, it should be borne in mind that political openness varies a great deal from one country to another. A large-scale survey may be dangerous to the liberty of researchers in some countries. 

Moreover, this study may be no worse than much-touted surveys showing Singapore being tops in economic attractiveness or having a fair judicial process. Or Changi Airport being number 1 in this and that. Those surveys often interview samples of businessmen and international travellers, but again, how are they selected?

As the Overview chapter in this book said, 

Asking the foreign business community about the political openness of a country is vastly different than asking the local civil society. By relying on citizens living in the country for their views, the [Asian Democracy Index] ensures that its respondents are familiar with the systems that they are asked to rate.


Six categories

The 22 questions each allowed 5 possible responses: Strongly agree, Agree, Don't know, Disagree and Strongly disagree. The formula for converting the responses into a percentage score is given in the box at right.

The questions were grouped into 6 categories, the first of which was Civil rights. In this category, there were 3 questions:

1.1 People can openly question and discuss official policy without fear.

1.2 The government allows citizens to demonstrate peacefully.

1.3 Civic associations and political organisations can freely organise, mobilise and advocate their views.

The average score for the above 3 questions in each of the 16 countries surveyed can be seen from the chart below. Singapore's score was 17.3%. Oh dear.

In the next section, I will detail the answers to one of these 3 questions (#1.3), so you will get a better picture of how that score came about.

The second category concerned Elections and political processes, with 5 individual questions:

2.1 Voters can freely choose their preferred candidates/parties without interference.

2.2 There is free and fair competition among political parties.

2.3 The election authorities and election officials are independent, impartial, and effective in guaranteeing free and fair elections.

2.4 There is a legitimate, non-arbitrary, transparent process to amend the constitution and laws.

2.5 External interference is a significant problem in our political process.

The last one stumps me. Is there a typographical error in the book? Should it not have been a question in the negative? Otherwise how was it scored?

Anyway, here are the country ratings. Singapore's 36.5% leaves us third from bottom:

Governance and corruption contained 3 questions:

3.1 The government's decision-making is transparent

3.2 Elected officials and civil servants are held accountable for their actions. 

3.3 The government is responsive to the public interest.

Singapore got an average score of 39.0% for these 3 questions, putting us more or less average among the various countries surveyed:

The next category was Media, with 4 questions:

4.1 The media is free and independent form government or other sources

4.2 Journalists and the media present a diversity of political views

4.3 The media contributes positively to democracy

4.4 Journalists have adequate access to public information

Our average score for these 4 questions was 28.4%. 

Rule of law comprised 3 questions.

5.1 The judicial system effectively protects human rights and democratic principles.

5.2 Constitutional and legal arrangements guarantee democratic process in practice.

5.3 All military, police and security forces are subject to civilian control. 

In this category, Singapore's average score was 26.9%.

The sixth and last category was Participation and representation.

6.1 Political parties provide an effective avenue for citizens' participation in politics.

6.2 Political parties adequately represent the interests of their constituents.

6.3 Civil society organisations effectively promote the public interest.

6.4 the rights and interests of vulnerable and threatened groups are adequately represented in the political system.

Again, Singapore was second from last with 34.3%.



Formula for scoring

A "Strongly disagree" gets 0 points. 
"Disagree" gets 1 point.
"Don't know" gets 2 points.
"Agree" gets 3 points, and
"Strongly agree" gets 4 points.

The total points are divided by 400 to get the percentage score.

Example: if for a certain question,

12% strong disagree, then 12 x 0 = 0
20% disagree, then 20 x 1 = 20
3% don't know, then 3 x 2 = 6
41% agree, then 41 x 3 = 123
24% strongly agree, then 24 x 4 = 96

Total = 245 points.

Divided by 400, gives a 61.25% score.

* * *

The percentage score for a category is the average of the scores for the various questions within that category.


Details from four questions

The above charts show average scores for each category across the 16 countries. For reasons of space, it is impossible for me to go further into each of the 22 questions, but in order to give readers a glimpse of the supporting data, I shall touch on 4 particular questions -- chosen randomly just as examples -- comparing Singapore with Hong Kong, another city-state with a similar level of economy.

From the Civil rights category is this question:

As you can see, some four-fifths of Singaporeans either disagreed or strongly disagreed. In Hong Kong, almost the same proportion agreed or strongly agreed.

From the Elections and political processes category is this question:

Some 70% of Singaporeans either disagreed or strongly disagreed. Opinion in Hong Kong was much more mixed.

From the Media category:

Of course, this result doesn't surprise us. Again, some four-fifths of Singaporeans gave a negative answer. Hong Kongers felt almost the exact reverse about their media.

Lastly, an example from the Rule of law category:

Is there anything left for me to say?

In closing

As I indicated above, skepticism about these findings can arise from the variability of procedures among the various countries, and the way respondents were selected.

To some Singaporeans, the involvement of Chee Soon Juan in ARDA, which commissioned the study, would colour their assessment of the results, especially considering the relative lack of transparency about the procedural details on the Singapore side.

It is partly because I anticipated such reactions that I presented the 4 pie-charts. Were the opinion patterns represented by the pie-charts more or less what one would expect Singaporeans to say? If so, then the scorings that were derived from them would be valid.

Moreover, skepticism about Singapore's scores has to be balanced by the fact that the partner organisations and individuals conducting the surveys in other countries were also human rights activists. If one suspects the Singapore results to be unreliable on account of the involvement of an opposition politician and civil rights activist here, one has to ask: how did similar activists in other countries still come up with scores that are better than Singapore's? Weren't they equally motivated to make things look bad in their countries? Is Chee a blacker devil than all the others?

Rather than pick on these results, the best thing to do is to use an approach crucial to the scientific process -- to attempt to replicate them. I hope political science students in Singapore will take up the challenge of doing a survey based on exactly the same questions, but with more rigorous selection of respondents and with a larger sample size. Then we'll know.

Yawning Bread