Gallup, 39% of national power vs. 36%

< Gallup, 39% of national power vs. 36% of Democrats - Is the reversal of party approval ratings due to 'conservative oversampling'?>

1>.
In the National Statistical Survey (NBS), which was released on Thursday, January 16, the approval rating of political parties stood at 35 percent for the national power and 33 percent for the Democratic Party. This is the first time that a telephone interview turned the tables.

On January 17 (Friday), the Gallup Korea survey was announced. It came out with 39 percent of the national power and 36 percent of the Democratic Party. The Gallup Korea survey also turned the tables. What a surprise. (*[Attachment - Figure 1])

Why did the national power get ahead of the Democratic Party? Two main hypotheses are raised about this.

► First, it is the 'conservative and sampling hypothesis'.

► Second, it is the 'Hypothesis that Representative Lee Jae-myung likes or dislikes'. As people gradually accepted it as the beginning of the 'presidential election phase', it seems that the alertness against the Democratic Party of Korea and Representative Lee Jae-myung was reflected. Let's take a look at each of these.

Three graphs are attached. ① The trend of party approval ratings ② The trend of the ideological orientation of conservative/moderate/progressive groups ③ It shows the trend of party approval ratings by each ideological orientation. Please refer to the three graphs and read them. (Graphs ② and ③ were compiled by Slow News CEO Lee Jung-hwan.)

2>.
► First, let's look at the 'conservative and sampling hypothesis'.

In Gallup Korea's third week of January, 338 conservatives, 278 moderates, 262 progressives and 123 unknown/non-response.

Let's look at the change in the response rate of remuneration. In November, it was 27%. In December, it was 32%. In January, it was 33%. Interestingly, the "increase in response to remuneration" did not happen in January. It happened in December. Compared to November, it increased by 5 percentage points in December. Compared to December, January only increased by 1 percentage point. (*[Attachment-Figure 2])

Again, let's come back to the question. Is the conservative oversampling? The conservative oversampling theory has to think about three things.

First, the ideological orientation of conservative and progressive originally fluctuates. Polls only meet demographic requirements. Age, region, and gender are representative. Political party approval ratings and ideological orientations are changed by 'subjective answers'. Progress increases sometimes, and conservatives increase sometimes.

Second, in order to use the term 'oversampling', 'normal sampling' is required. Strictly speaking, there is no 'normal ratio' sampling. Why? Because it is supposed to change.

Third, we summarized the 'long-term average'. From July 2023 to January 2025, we calculated the long-term average of remuneration. First, the long-term average for 19 months of the conservatives was 27.5%.

Long-term average for 19 months of the 'conservative class' from July 2023 to the latest) 27.5%
Gallup Korea tallies 33.8% for 3rd week of January conservatives
The gap between the two is 6.3 percentage points.

Compared to the 'long-term average' for 19 months, it is true that the remuneration has increased by 6.3 percentage points. However, what is interesting is that there have been more responses to remuneration from 'December', not from January.

3>.
► Second, let's look at the likes/dislikes hypothesis about representative Lee Jae-myung. This hypothesis believes that the Yoon Suk Yeol referee was the main concern in the 'impeachment phase', and the likes/dislikes of representative Lee Jae-myung shifted to the 'presidential phase'.

It is a view that the pros and cons of representative Lee Jae-myung appear in the form of a 'party approval rating'. One of the ways to know this is by looking at 'excluding the conservatives' and 'the change in public opinion among the middle class'.

[Attachment-Figure 3] summarizes the changes in party approval ratings of progressives, moderates, and conservatives in the second week of December, the second week of January, and the third week of January, respectively.

Among them, only the "ideological middle class" changes. In the second week of December (just before the impeachment motion was passed by the National Assembly), the Democratic Party (36 percent) and the national power (19 percent) accounted for the moderate ideological group. The gap between the two was 17 percentage points.

However, in the third week of January, the Democratic Party and the national power accounted for 37 percent and 28 percent. The gap between the two was 9 percentage points.

Compared to the middle-of-the-road public opinion in the second week of December and the third week of January, support for the Democratic Party increased by 1 percentage point. Support for national power increased by 9 percentage points. The middle class is the largest group. On a 19-month average, it accounts for 42.7% of all voters.

4>.
Let's sort it out.

First, in the Gallup Korea survey, the party's approval rating of 'national power' is rising. It is true that there are more <conservative sampling> than the long-term average. Whether this is described as 'conservative assembly' or 'oversampling of conservatism', both are correct. However, that much is a change of about 6.3 percentage points in the total voter.

Second, looking at the change in the approval rating of the <middle class>, the Democratic Party is in place, and the national power has risen relatively.

What does this suggest for the presidential election? Does the national power win, and the Democratic Party loses? No, because other figures still favor the Democratic Party.

Two figures are in favor of Democrats. For and for impeachment. 56% approve of impeachment vs. 36% disapprove of impeachment. (*This gap is greatly diluted if there is a "yes to impeachment" candidate from national power.)

The result is whether they approve of the change of government. 48 percent of the respondents said "change of government" vs. 40 percent said "re-creation of government." Public opinion on the change of government is 8 percentage points higher.

However, the approval rating of political parties is 39 percent for national power vs. 36 percent for the Democratic Party. There is a possibility that there is a "conservative taxation" that is around 6 percentage points. That is a good point.

However, it cannot be explained by conservative taxation alone. This is because the gap in party approval ratings is narrowing even in the 'middle class'.

5>.
So, what should be done? Elections are 51 percent games. Fifty-one percent are realized only through a coalition of voters.

The Democratic Party of Korea and representative Lee Jae-myung need to operate an action plan to appeal to the middle class. It is necessary to prepare both a policy response and a political response at the same time. (There will be a limit to policy responses alone.)

Of course, analysis and prospects can be wrong. However, you should always be wary of the attitude of only looking at what you want to see.

Denying the election results will lead to a conspiracy theory of fraudulent elections.
If you deny the results of the poll, you will fall into the <poll conspiracy theory>.

I don't think poll company A is unconditionally right, and poll company B is unconditionally wrong. If you think in a balanced way, look at the time series trend of 'the same company', but several companies at the same time