Sample size of one hardly any better than biased samples

I refer to the 17 Jun 2014 Straits Times letter “Don’t take good govt for granted” by Mr Eugene Tan.

Mr Tan ridiculed Ms Catherine Lim for her silence on Singapore’s supposed near top ranking in Gallup’s 2013 Most Emotional Society survey when in the previous year Ms Lim attributed Singapore’s bottommost ranking in the 2012 edition of the same survey to authoritarian government policies. Mr Tan is wrong. Singapore came in 67th of 143 nations in the 2013 survey, putting Singapore at the 47th percentile, hardly the so called “appearing near the top of the list” as claimed by Mr Tan. Let’s repeat this to Mr Tan: 67th of 143 is far from, not near the top.

• Singapore came in 67th out of 143 countries in the “positive experience” index in the latest survey, which was carried out last year. The findings were released yesterday.
In the 2011 study, it was the least positive of 148 countries – its worst ranking since 2007, when it first took part in the study.
[Straits Times, Method behind the survey, 1 Oct 2013]

Mr Tan contrasted PAP’s low percentage votes of 61% and 65% respectively during the non-crisis years of 1992 and 1997 with its high percentage votes of 75.3% during the crisis year of 2001 to show that when it comes to the crunch, people trust PAP.

The following table shows that 2001 was the only year when a crisis year was also an election year. Hence, Mr Tan’s theory that crisis years means good election years for PAP has a miserable sample size of just one only, hardly sufficient for making anything out of.

Year PAP election vote percentage PAP leader Year Recession events
1968 86.7% Lee Kuan Yew
1972 70.4% Lee Kuan Yew
1976 74.1% Lee Kuan Yew
1980 77.7% Lee Kuan Yew
1984 64.8% Lee Kuan Yew
Lee Kuan Yew 1985 Singapore’s first ever recession
1988 63.2% Lee Kuan Yew
1991 61.0% Goh Chok Tong
1997 65.0% Goh Chok Tong
Goh Chok Tong 1998 Asian Financial Crisis
2001 75.3% Goh Chok Tong 2001 Sept 11 Terrorist attacks
2006 66.6% Lee Hsien Loong
Lee Hsien Loong 2009 Global Financial Crisis
2011 60.1% Lee Hsien Loong

While having the cheek to wax lyrical about the need to avoid sampling bias when quoting the opinion poll taken before the 1948 United States presidential election, Mr Tan ended up making a statement out of a sample size of just one only. Does Mr Tan not realize that using a sample size of only one is hardly any better than using biased samples?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: