More than just "same story, different emphasis".....

[ncr]

These two news items by
MCOT/TNA (Poll finds majority want PM, main critic to reach compromise) and
The Nation (Next anti-Thaksin rally to be bigger: survey) actually constitute quite an extreme example of discrepancy.

Originally Posted by MCOT

Despite the recent political upheaval and allegations directed at Mr. Thaksin, the poll found his approval rating had increased sharply from 33 per cent to 59 per cent.

-contrast that with:-

Originally Posted by The Nation

The poll found almost 55 per cent of respondents continue to trust the premier's administration of the country. That was a drop of 4.4 percentage points over the previously poll.

Not sure if they are referring to the same question, though.

And another case of confusion (what was the original poll question?):

Originally Posted by MCOT

Half of all respondents wanted to learn more about the prime minister, 25 per cent showed an interest in Industry Minister Suriya Jungrungreangkit, and 15 per cent Agriculture and Cooperatives Minister Khunying Sudarat Keyuraphan.

Originally Posted by The Nation

When asked which ministers will be named by Sondhi tomorrow, more than 50 per cent said Prime Minister Thaksin, 25.5 per cent said Industry Minister Suriya Jung-rungreangkit and 15.3 per cent said Agriculture Minister Khunying Sudarat Keyuraphan.

I don't understand a word, anyway. "...wanted to learn more about"? "...will be named by Sondhi"? What the heck is that supposed to mean? Information value = close to zero.

So with respect to the first two quotes: who is twisting the truth here? Creeping suspicion: both. (Which only serves as a reminder that you always have to be careful believing what's being told in the media......)

[BangkokPundit]

NCR

Here is the poll in question. I noticed something similiar as well, between the Nation and article in another newspaper (BKK Post ?). But by the time I found the poll, I couldn't find the second article!

If I have time, I will do a translation of parts of the poll in the next couple of days.

[BangkokPundit]

The poll in question is here.

Originally Posted by MCOT
Despite the recent political upheaval and allegations directed at Mr. Thaksin, the poll found his approval rating had increased sharply from 33 per cent to 59 per cent.

Quote:
Originally Posted by The Nation
The poll found almost 55 per cent of respondents continue to trust the premier's administration of the country. That was a drop of 4.4 percentage points over the previously poll.

There are 3 polls. The poll of 3-4 Feb, shows that 33.6% of people want Thaksin to continue governing. This increased to 59.3% for the poll of 5-6 February, but dropped to 54.9% for the latest poll (7-8 Feb). This time, the Nation is correct. The MCOT are correct that the number of people who want Thaksin to continue to govern did increase from 33% to 59% but they left the results from the latest poll.

The MOCT article was published on 9 February and the lead says they are referring to the latest poll and that the poll they are talking about surveyed 1800 people. The poll of 7-8 Feb surveyed 1798 people whereas the poll of 5-6 Feb surveyed only 1377 people. The MCOT article is misleading at best.

Originally Posted by MCOT
Half of all respondents wanted to learn more about the prime minister, 25 per cent showed an interest in Industry Minister Suriya Jungrungreangkit, and 15 per cent Agriculture and Cooperatives Minister Khunying Sudarat Keyuraphan.

Originally Posted by The Nation
When asked which ministers will be named by Sondhi tomorrow, more than 50 per cent said Prime Minister Thaksin, 25.5 per cent said Industry Minister Suriya Jung-rungreangkit and 15.3 per cent said Agriculture Minister Khunying Sudarat Keyuraphan.

The Thai for this part is rather difficult to translate.

For the cabinet ministers who people want to hear Sondhi name (because they want to hear more dirt on them ?), the top 3 are Thaksin, Sudarat, and Suriya.

The MCOT's version sounds much closer. The Nation has just got it completely wrong.

[Scuba22]

Whoa whoa, hang on here - and forgive me since I can't read thai, but in English there's a HUGE difference between "approval", "trust in administration", and "want to continue to lead the country".

I can definitely see where people may not want to change a leader even if they don't "approve" of what they're doing - the 2004 elections in the US are a great case in point. In turbulent times, you may hold your nose and go for stability even though you hate the bastard.

A swing of over 25 points within 3 days on a single measure in a comparable sample is really hard to imagine without an extraordiary event intervening, like a terrorist attack or a currency devaluation. I wouldn't expect any mass rally to have an impact like that unless there was serious violence.

A swing from 33% to 59% definitely sounds like different measures - are you sure you got the questions correct?

Cheers,

Scuba22

[BangkokPundit]

Originally Posted by Scuba22

Whoa whoa, hang on here - and forgive me since I can't read thai, but in English there's a HUGE difference between "approval", "trust in administration", and "want to continue to lead the country".

Yes, I agree. I was going to mention this, but I found myself getting into a semantic debate on what the exactly the wording should be and well pointing out all the errors in The Nation and MCOT is more than a 1 person job.

First, the survey for 3-4 Feb is here.

The relevant part is:

... 4 ก.พ. 49 นั้นพบว่า ตัวอย่างร้อยละ 33.6 ให้ทำงานต่อ ร้อยละ 14.6 ให้ลาออก ร้อยละ 7.2 ให้ยุบสภา ร้อยละ 6.2 ระบุอื่นๆ

... on 4 February 2006 found that 33.6% of the sample size want [the PM] to continue to work, 14.6% want [the PM] to resign, 7.2% want [the PM] to dissolve parliament, [and] 6.2% want [the PM] selected other.

Second, the later survey, which the articles are talking about.

The Thai is below:

... 3-4 ก.พ. เปรียบเทียบกับการสำรวจในครั้งนี้ โดยพบว่า สัดส่วนของตัวอย่างที่ระบุต้องการให้นายกรัฐมนตรีทำงานต่อไปนั้นเพิ่มขึ้นจากร้อยละ 33.6 ในการสำรวจครั้งก่อน เป็นร้อยละ 59.3 ในการสำรวจเมื่อวันที่ 5-6 ก.พ. และมีแนวโน้มลดลงเล็กน้อยในการสำรวจครั้งล่าสุด (ร้อยละ 54.9)

... on 3-4 Feb compared with this survey found that the proportion of the sample who specified that they want the PM to continue to work increased from 33.6% from the previous survey to 59.3% for the survey on 5-6 of Feb and there is a downward trend in the latest survey (54.9%).

Originally Posted by Scuba22

A swing of over 25 points within 3 days on a single measure in a comparable sample is really hard to imagine without an extraordiary event intervening, like a terrorist attack or a currency devaluation. I wouldn't expect any mass rally to have an impact like that unless there was serious violence.

A swing from 33% to 59% definitely sounds like different measures - are you sure you got the questions correct?

I don't disagree. The the later survey, which the articles are talking about says this:

ซึ่งผลการวิเคราะห์พบว่าส่วนต่างที่เพิ่มขึ้นดังกล่า วนี้น่าจะมาจากการเปลี่ยนแปลงทางความคิดของตัวอย่างท ี่ไม่ระบุความคิดเห็นในการสำรวจครั้งก่อนหน้านี ้ เป็นสำคัญ

I found this difficult to translate, but a rough translation is 'results of the analysis shows that the likely reason for the difference between the surveys comes from a change of thinking in people who in the previous survey did not specify an opinion [I assume this means they took a wait and see aproach and were 'undecided']'

This makes it sound much more plausible so instead of people changing their mind, they went from undecided to wanting the PM to continue to govern. To be honest, I am not surprised, if I listened to Sondhi and Co harp on, I am not sure that I would really see that there is any viable alternative to Thaksin - hence why people still want him to govern.

Back to your initial question, the wording of the poll is certainly not "trust in administration". I think "approval" is just as inaccurate as well. The Nation and the MCOT both do badly on this part as well. The most accurate is continue to work/continuing to work. "Lead" is not the word used in Thai.

[The Enforcer!]

Of course the most obvious reason why MCOT results differ from anyone else is that they are a government department.

The Enforcer!