Machine Learning Algorithms for Big Data Applications With Policy Implementation

This article examines the policy implementation literature using a text mining technique, known as a structural topic model (STM), to conduct a comprehensive analysis of 547 articles published by 11 major journals between 2000 and 2019. The subject analyzed was the policy implementation literature, and the search included titles, keywords, and abstracts. The application of the STM not only allowed us to provide snapshots of different research topics and variation across covariates but also let us track the evolution and influence of topics over time. Examining the policy implementation literature has contributed to the understanding of public policy areas; the authors also provided recommendations for future studies in policy implementation.

Management and Organizational Research: Structural Topic Modeling for a Better Understanding of Theory Application

In management and organization research, theory development is often linked with developing a new theory. However, regardless of the number of existing theories, most theories remain empirically untested, and the progress in understanding the application of theories has been scarce. This article discusses how theories are applied in existing management and organization research studies. This study applies the Structural Topic Model to 4636 research papers from the S2ORC dataset. The results reveal twelve research themes, establish correlations, and document the evolution of themes over time. The findings of this study reveal that the theoretical application is not consistent across research themes, theories are primarily used for descriptive and communicative properties, and most research themes in management and organization research are more concerned with discovering phenomena rather than with understanding and forecasting them.

Language in Congress: Domestic Constituent Influence in Foreign Aid Decisions

There is an extensive literature on the effect of donor ideology on foreign aid allocations. However, the process through which donor ideology influences aid decisions is understudied. In my framework, legislators' application of political ideology is expanded to foreign aid agendas through interactions with domestic constituencies: development Non-Governmental Organizations (NGOs) and private enterprises. Legislators adopt the constituencies' ideological rationale for aid and reflect the groups' aid preferences by taking on the language of those constituents. To test this argument, I applied the Structural Topic Model (STM) and Wordfish to my self-collected text data on testimonial statements given by representatives of NGOs and of firms and floor speeches of left- and right-leaning legislators relating to foreign aid in Congress. My results suggest that constituent groups have an influence on the ideological aid positions of individual legislators, which, in turn, may translate into the aid decisions of the donor country.

What Are the Salient and Memorable Green-Restaurant Attributes? Capturing Customer Perceptions From User-Generated Content

This study aims to survey user-generated content (UGC) from diners in certified green restaurants, discover the green images they recall, and demonstrate the usefulness of applying a probabilistic topic model to comprehend customers’ perceptions. Postvisit online reviews ( N = 28,098), in the form of unstructured texts from the TripAdvisor.com website, were used to find freely recalled green-restaurant images. These data were preprocessed with a structural topic model (STM) algorithm to select 51 relevant categories of images. These image categories were compared with the findings of previous studies to discover unique restaurant attributes. Furthermore, a topic-level network and a green-restaurant network were drawn to discover the most easily recallable image categories and their attributes. This machine-learning-based approach improved the reproducibility of unstructured data analyses, overcoming the subjectivity of qualitative data analysis. Theoretical and practical implications are offered for topic modeling methodology along with marketing strategies for restaurateurs.

Structural Topic Model Analysis of Mask-Wearing Issue Using International News Big Data

Media plays an important role in the acquisition of health information worldwide. This was particularly evident in the face of the COVID-19 epidemic. Relatedly, it is practical and desirable for people to wear masks for health, fashion, and religious regions. However, depending on cultural differences, people naturally accept wearing a mask, or they look upon it negatively. In 2020, the COVID-19 pandemic led to widespread mask-wearing mandates worldwide. In the case of COVID-19, wearing a mask is strongly recommended, so by analyzing the news data before and after the spread of the epidemic, it is possible to see how the direction of crisis management is being structured. In particular, by utilizing big data analysis of international news data, discourses around the world can be analyzed more deeply. This study collected and analyzed 58,061 international news items related to mask-wearing from 1 January 2019 to 31 December 2020. The collected dataset was compared before and after the World Health Organization’s pandemic declaration by applying structural topic model analysis. The results revealed that prior to the declaration, issues related to the COVID-19 outbreak were emphasized, but afterward, issues related to movement restrictions, quarantine management, and local economic impacts emerged.

Measuring Equity-Promoting Behaviors in Digital Teaching Simulations: A Topic Modeling Approach

Diversity, equity, and inclusion (DEI) issues are urgent in education. We developed and evaluated a massive open online course ( N = 963) with embedded equity simulations that attempted to equip educators with equity teaching practices. Applying a structural topic model (STM)—a type of natural language processing (NLP)—we examined how participants with different equity attitudes responded in simulations. Over a sequence of four simulations, the simulation behavior of participants with less equitable beliefs converged to be more similar with the simulated behavior of participants with more equitable beliefs ( ES [effect size] = 1.08 SD). This finding was corroborated by overall changes in equity mindsets ( ES = 0.88 SD) and changed in self-reported equity-promoting practices ( ES = 0.32 SD). Digital simulations when combined with NLP offer a compelling approach to both teaching about DEI topics and formatively assessing learner behavior in large-scale learning environments.