□ Public deliberation has demonstrated its potential as a conflict resolution mechanism through deliberative discourses on Shin-Kori nuclear power plant units 5 and 6 and the college admissions reform. However, these deliberative processes revealed some flaws in representativeness, inclusiveness, and the derivation of social consensus building through deliberation. In light of growing efforts to incorporate it into public decision-making, there is also an increasing need for proficiencies in systematic planning, support, and oversight.

- Public deliberation on Shin-Kori NPP (units 5 and 6) in 2017 showed its potential as a new mechanism for conflict resolution.

- In a public survey on the wide use of public deliberation, 75.9% are satisfied while a mere 4.7% remain worried, but still, there are both expectations and concerns.

- Discussions on the pros and cons of public deliberation need to be based on objective evaluations of respective cases.

- Deliberative democracy is a complementary system to representative democracy, and unlike participatory democracy, it undertakes a debate process to check on public opinions and attempts to make decisions based on the confirmed opinion.

- Deliberative democracy gives legitimacy to the decision made by ordinary citizens and their representatives through participation and deliberation.

- Public deliberation can be assessed and evaluated under three frameworks of planning, participation, and deliberation.

- Citizen participants must be able to fully represent and include the general public who will be affected by their collective decisions and policies.

- A proportional extraction of population is not enough to fully ensure the representativeness of deliberative discourses (principle of inclusiveness).

- Preliminary public surveys and deliberation groups showed a statistically meaningful difference regarding conflicting policy issues.

- In terms of the composition of citizen participants, supporters of nuclear phaseout in the Shin-Kori case and those prioritizing school records in the college admissions case were overrepresented.

- Among preliminary survey respondents who had expressed their willingness to participate in public deliberation, citizen participation groups were selected using the stratified extraction method.

- Politically apathetic or conservative citizens tend to be systematically less willing to participate in public deliberation.

- By the criteria of inclusiveness, the final sampling of participants was biased towards amplifying the shortcomings of representativeness.

- The deliberation group for Shin-Kori NPP Units 5 and 6 got more and more right answers in knowledge quizzes with each survey.

- Valuing safety and the environment tipped the scales toward construction suspension, and stressing economic and industrial aspects tilted it to resumption.

- A shift in preference among debate participants most strongly demonstrates reflective deliberation.

- In the deliberative process, 32.7% formed opinions, 7.0% changed viewpoints, and 58.6% maintained initial positions.

- After public deliberation, the Shin-Kori NPP construction transformed from a contest between 29% for suspension vs. 37% for resumption to a debate between 41% vs. 56%.

- Knowledge gain functioning as the primary driver of preference shifts supports that such changes are more likely when the level of understanding improves.

- In the college admissions reform deliberation, the completion rate of online lectures by debate participants surpassed 97%, and the number of correct answers to knowledge questions gradually increased over time throughout the deliberative process.

- A factor analysis narrowed down eight opinion bases into two perspectives: ① fair opportunities in education and ② efficiency of the education system.

- The tendencies became clear to favor Agenda 1 and 4 as more supportive of fair opportunities in education and Agenda 2 and 3 with a higher value on efficiency and safety of the overall education system

- In the college admissions deliberation, twothirds of participants experienced a shift in preference.

- The college admissions reform case also shows that knowledge gain increased the probability of preference shift.

- The factor analysis extracted a single common factor (metaissue) that runs through four agendas.

- In light of the correlation with each agenda, the meta-issue translates as preferences for CSAT.

- Potential preferences for CSAT tend to be more scattered after the deliberative process.

- The deliberative process might have reinforced the divide among participants rather than closing it, and perhaps agendasetting was not properly designed to capture key agenda issues.

- Regarding participation, the criteria of demographic representation were satisfied, but the discourses were neither representative nor inclusive enough.

- As for deliberation, the analysis confirmed that the better participants understand the debate issue, the more likely they form a reasonable opinion.

- However, public deliberation came short of building a social consensus through its process.

- A deeper analysis is necessary as to what caused the dispersion of viewpoints in the case of college admissions reform.

- The design of the admissions reform debate was unnecessarily complex.

- Without a step closer to public opinion formation, public deliberation is just another form of majority rule with some deliberation process.

- Now is the time to build resources, capacities, and institutional mechanisms for more systematically designed and managed public deliberation.