As the race for next year's presidential election heats up day by day, the public is being bombarded by one or more polls every week.
How do we interpret these poll results? Does public opinion really fluctuate with more unpredictability than the tide?
I believe that the people from the Lien Chan (
Strangely enough, after the James Soong (
However, another poll conducted by TVBS and released the same evening completely contradicted those results. According to the TVBS survey, the support rates for Lien, Soong and Chen were 18 percent, 27 percent and 31 per-cent, respectively
What happened? Although the two polls were conducted one to two days apart, I do not believe public opinion changes so rapidly.
To eliminate such confusing and contradictory results, I suggest that those conducting polls focus on establishing credibility and eliminating the margin of error in their statistics both to protect the quality of their sampling survey and prevent the results from being abused.
The error of statistical inference in surveys is an important element often overlooked by many people when reading poll results. A poll conductor would say Chen ranks above Lien when Lien's support rate is 20.3 percent and Chen's 21.4 percent. Actually, this is inaccurate. It is more accurate to say that with a 95 percent confidence level, Lien's support rate is between 17.87 and 22.73 percent and Chen's support rate is between 18.93 and 23.87 percent, out of 1,056 samples and assuming that no non-sampling error exists.
Based on this, who ranks ahead of the other? It is hard to tell. Determining ranking based on one single statistical figure is meaningless and misleading.
Recently, China Television (
Does this explanation sound reasonable? It is far from reality.
We know that AC Nielsen's viewership polls have a statistical margin of error. When AC Nielsen estimated that the viewership rating for China Television news was 6.62 percent and 6.47 percent for Taiwan Television (
What AC Nielsen does is rank China Television on top, with Tai-wan Television in second. It is, in fact, providing inaccurate rank-ings. Furthermore, AC Nielsen actually places news programs and variety shows on the same ranking chart. This is as absurd as comparing apples and oranges.
AC Nielsen is a famous international research and marketing consulting firm. It should not offer these less than professional statements. Perhaps it could indicate that the rankings are for reference only and should not be taken seriously. However, what is the point of an inconclusive analysis?
What we hope the company would do is accurately describe the margin of error in their ranking chart to keep others from deliberately misusing or misreading the chart. Otherwise, we ask that the company act professionally and tell the media that the viewership ratings are simply an estimate, with a real possibility for error, and the media should not rank ratings based directly on the rating chart. If AC Nielsen is contracted to provide a ranking chart, then it must provide a more statistically accurate ranking.
Statistical sampling should be conducted in a scientific manner. Hopefully, the pollsters would sample scientifically to ensure the quality of the statistical data gathered. Even more importantly, I hope that all pollsters would analyze popular polls and market surveys in a scientific manner. The results of the polls should not be over-simplified and the conclusions reached should not be ambiguous or misleading.
Cheng Kuang-fu is a professor at the Graduate Institute of Statistics at the National Central University.
The Chinese government on March 29 sent shock waves through the Tibetan Buddhist community by announcing the untimely death of one of its most revered spiritual figures, Hungkar Dorje Rinpoche. His sudden passing in Vietnam raised widespread suspicion and concern among his followers, who demanded an investigation. International human rights organization Human Rights Watch joined their call and urged a thorough investigation into his death, highlighting the potential involvement of the Chinese government. At just 56 years old, Rinpoche was influential not only as a spiritual leader, but also for his steadfast efforts to preserve and promote Tibetan identity and cultural
Former minister of culture Lung Ying-tai (龍應台) has long wielded influence through the power of words. Her articles once served as a moral compass for a society in transition. However, as her April 1 guest article in the New York Times, “The Clock Is Ticking for Taiwan,” makes all too clear, even celebrated prose can mislead when romanticism clouds political judgement. Lung crafts a narrative that is less an analysis of Taiwan’s geopolitical reality than an exercise in wistful nostalgia. As political scientists and international relations academics, we believe it is crucial to correct the misconceptions embedded in her article,
Strategic thinker Carl von Clausewitz has said that “war is politics by other means,” while investment guru Warren Buffett has said that “tariffs are an act of war.” Both aphorisms apply to China, which has long been engaged in a multifront political, economic and informational war against the US and the rest of the West. Kinetically also, China has launched the early stages of actual global conflict with its threats and aggressive moves against Taiwan, the Philippines and Japan, and its support for North Korea’s reckless actions against South Korea that could reignite the Korean War. Former US presidents Barack Obama
The pan-blue camp in the era after the rule of the two Chiangs — former presidents Chiang Kai-shek (蔣介石) and Chiang Ching-kuo (蔣經國) — can be roughly divided into two main factions: the “true blue,” who insist on opposing communism to protect the Republic of China (ROC), and the “red-blue,” who completely reject the current government and would rather collude with the Chinese Communist Party (CCP) to control Taiwan. The families of the former group suffered brutally under the hands of communist thugs in China. They know the CPP well and harbor a deep hatred for it — the two
RSS