Korea Information Society Development Institute

This study investigates the trends and peculiarities of doctoral research in the Republic of Moldova through an analysis of bibliographical references in doctoral theses from 2005 and 2023. Aiming to understand the features and evolution of citation practices, the study examines 375 theses available in national databases. The research addresses the quantity, nature, and language of citations, revealing significant findings. It shows an increase in the average number of references per thesis over time, indicating heightened scientific rigor and comprehensive literature integration. Differences across scientific fields are noted, with history and political science theses having the most references and computer science and physical-mathematical sciences the fewest. The analysis finds that scientific publications dominate the references, though other sources are important in some socio-humanities fields. Language trends indicate a shift from Russian to English references, reflecting increased global integration. The study also highlights the correlation between the thesis language and the predominant language of references used. These insights can be useful for policymakers, researchers, and academic institutions in understanding and improving the quality and impact of doctoral research in Moldova.

Since the launch of ChatGPT in November 2022, the number of services providing generative AI has been steadily increasing. As different services enter the market, the generative AI interfaces users experience become more diverse. However, none of the services has yet established itself as the dominant tool, and which interface component affects user preferences the most has yet to be identified. This study investigates user preferences based on the interface of generative AI services currently on the market. We investigated users’ preferred interface components by setting the output data type, generative style, output variation, reference style, and generation history provided by the current generative AI service as properties. We collected data from 500 users through a survey and conducted conjoint analysis. Users preferred the provision of 10 generation histories the most, and the second most preferred the provision of reference style in footnote format. In addition, it was found that there was no preference for the creative generative style, which can be interpreted as users being aware of the problem of hallucination in generative AI. The results of this study will help future generative AI services design interfaces that consider user experience.

Wireless emergency alerts (WEAs), which deliver disaster information directly to individuals' mobile phones, have been widely used to provide information related to COVID-19 and to encourage compliance with social distancing guidelines during the COVID-19 pandemic. The floating population refers to the number of people temporarily staying in a specific area, and this demographic data can be a useful indicator to understand the level of social distancing people are complying with during the COVID-19 pandemic. This study aimed to empirically analyze the impact of WEAs on the floating population where WEAs were transmitted in the early stages of the COVID-19 pandemic. As most WEA messages focus on compliance with the government's social distancing guidelines, one of the goals of transmitting WEAs during the COVID-19 pandemic is to control the floating population at an appropriate level. We investigated the empirical impact of WEAs on the floating population across 25 districts in Seoul by estimating a panel regression model at the district-hour level with a series of fixed effects. The main independent variables were the number of instant WEAs, the daily cumulative number of WEAs, the total cumulative number of WEAs, and information extracted from WEAs by natural language processing at the district-hour level. The data set provided a highly informative empirical setting as WEAs were sent by different local governments with various identifiable district-hour-level data. The estimates of the impact of WEAs on the floating population were significantly negative (-0.013, P=.02 to -0.014, P=.01) across all specifications, implying that an additional WEA issuance reduced the floating population by 1.3% (=100(1-e-0.013)) to 1.4% (=100(1-e-0.014)). Although the coefficients of DCN (the daily cumulative number of WEAs) were also negative (-0.0034, P=.34 to -0.0052, P=.15) across all models, they were not significant. The impact of WEAs on the floating population doubled (-0.025, P=.02 to -0.033, P=.005) when the first 82 days of observations were used as subsamples to reduce the possibility of people blocking WEAs. Our results suggest that issuing WEAs and distributing information related to COVID-19 to a specific district was associated with a decrease in the floating population of that district. Furthermore, among the various types of information in the WEAs, location information was the only significant type of information that was related to a decrease in the floating population. This study makes important contributions. First, this study measured the impact of WEAs in a highly informative empirical setting. Second, this study adds to the existing literature on the mechanisms by which WEAs can affect public response. Lastly, this study has important implications for making optimal WEAs and suggests that location information should be included.

<p>The deliverable contains the result of the analysis and benchmarking regarding the institutional Open Access policies, and presents a support for the preparation of the implementation stage of the Open Science policies in the Republic of Moldova.</p>

The deliverable presents the results of the survey "Attitudes towards Open Science", carried out within the project "Boosting engagement of the Republic of Moldova in Open Science: methodological and applied support" (cod 21.70105.40ȘD), between October 14 and December 6, 2021. 685 researchers, scientific and didactic staff, PhD students, librarians, R&D decision makers and other stakeholders participated in the survey. The final sample comprised n = 532 participants. The highest participation rate in the survey (79.89%) belongs to representatives of higher education institutions, followed by research institutions (16.54%). The survey participants represent all fields of research, about half are from the field of social and economic sciences (26.88%) and humanities (21.81%). The survey showed that 85.15% of respondents are familiar with the concept and practices of Open Science, and more than 50% of participants practice and promote Open Science. Among the survey participants, only 14.10% of respondents are not at all familiar with the concept and practices of Open Science. The survey data ascertains the motivation from the scientific community regarding the practice of Open Science, especially regarding Open Access to publications, open educational resources, increased visibility, access to research infrastructures, etc. At the same time, several problems and concerns were noted, especially related to the correct use of research data, copyright, abusive use of research results, lower quality, etc. The participants of the survey believe that the existence of a Policy or Strategy at the national level regarding Open Science is important (95.13%). The most important key aspects to be considered by a National Open Science Policy or Strategy that were selected by the respondents are the same as for an institutional policy or strategy: Open Access to Publications, Open Access to Research Data and Access open to educational resources. The majority of survey participants (67.85%) believe that Open Science is an opportunity for science.

Abstract Export and productivity growth in the Korean manufacturing sector have both been slow since 2011. To understand this relationship, we examine the productivity distributions of manufacturing firms in Korea by applying the dynamic selection model developed by Sampson (2016). The fitness of the dynamic selection model suggests that a slowdown in the market selection mechanism may be a crucial contributor to the stagnant growth in the productivity. In addition, we find that the productivity difference between exporting firms and non‐exporting firms has decreased and the low productivity growth of firms exporting to non‐foreign affiliates has contributed to this trend.