UROP Research Mentor Project Submission Portal: Submission #497

Submission information
Submission Number: 497
Submission ID: 8756
Submission UUID: 0e1d6ca7-9dda-4029-975e-dbd08e0fbcf4

Created: Sat, 08/19/2023 - 10:46 PM
Completed: Sat, 08/19/2023 - 11:00 PM
Changed: Fri, 09/29/2023 - 10:29 AM

Remote IP address: 98.230.43.233
Submitted by: Anonymous
Language: English

Is draft: No

Research Mentor Information

Guimin Zheng
She/Her
Ms.
gzheng@fsu.edu
Graduate Student
{Empty}
{Empty}
Social Sciences and Public Policy
Askew School of Public Administration
Guimin Zheng.JPG

Additional Research Mentor(s)

Jing He
She/Her
Ms.
jhe@fsu.edu
{Empty}
{Empty}
{Empty}
{Empty}

Overall Project Details

Unmasking Public Leadership through Tweets: A Machine Learning Analysis of U.S. City Officials’ Social Media Use
public leadership, social media use, local government officials, machine learning methods, twitter
No
2
Open to all majors, but prefer students from all social science majors, statistics, and computer science.
On FSU Main Campus
No, the project is remote
Fully Remote
8
During business hours
Leadership is among the core concepts in public administration (PA) (Chapman et al., 2016; Crosby & Bryson, 2018; Ospina, 2017; Van Wart, 2013). Public leadership is a phenomenon worth studying because, in an increasingly complex and ambiguous world, new challenges and pressures are placed on public organizations and their leaders (Van Wart, 2013). A growing body of literature has examined transformational leadership, but many more public leadership styles have recently been identified yet underexamined. Moreover, while many studies measure public leadership through interviews and survey questions, which is quite helpful in obtaining first-hand perspectives, there is a lack of objective measures. Therefore, this study aims to bridge the gap by facilitating a comprehensive grasp of how different types of public leadership orientations are manifested by local governmental leaders within the succinct medium of Tweets. Relying on posts published by verified city leaders’ Twitter accounts, we use BERTweet, a pre-trained machine learning language model for English Tweets, to conduct a partly automated content analysis to elucidate distinct leadership patterns exhibited within these digital communications.

Specifically, we will focus on four public leadership styles that government leaders actively use in dealing with public sector issues: (1) accountability leadership, (2) rule-following leadership, (3) political loyalty leadership, and (4) network governance leadership. First, accountability leadership promotes dialogue and encourages employees to justify their actions to their wider stakeholders, including politicians, citizens, and nongovernmental organizations (Roberts 2003). Accountable leaders encourage employees to be open and honest with their internal and external stakeholders and keep them informed of their progress and decisions. Second, rule-following leaders encourage rule-driven behavior that is trans-situational and not actor-specific (Klijn and Koppenjan 2016), urge employees to adhere to government rules and regulations, and take active steps to ensure that these rules and regulations are followed (Tummers and Knies 2016). Third, political loyalty leadership is defined as motivating employees to follow through on politicians’ decisions, even when they are costly to them. Leaders with high levels of political loyalty encourage and reward employees for following through on policy directives, even when they personally disagree and or when a bill negatively affects their own department (Heidari-Robinson 2017). Finally, network governance leadership is shown when public leaders actively encourage employees to network and connect with various stakeholders in their own organizations and in the wider community (Tummers and Knies 2016). These leaders expand employees’ network and knowledge of who their stakeholders actually are, which helps employees to put a face to the departments, agencies, and communities they serve.

For future agendas, there will be more exciting research ideas and questions we can explore with the Twitter data we collect.
(1) Student research assistants will help manually collect social media accounts data, including the Twitter and Facebook accounts of about 1,200 incumbent mayors, public managers, and/ or city councilors, based on a full name list we already have. These city leaders come from over 200 U.S. cities with populations of 30,000 to over 500,000, according to the 2020 U.S. Census.
(2) They will do the human annotation tasks on the scraped posts. High-quality annotated data is very important to the later machine learning training and testing analysis.
(3) If possible, they will read relevant documents, news, and journal articles to understand how researchers use Twitter data to conduct research and share their new ideas and thoughts on interesting research questions or topics with Twitter data.
(4) If time permits, they will help analyze the data with us and interpret the statistical results.
Required:
- Use Microsoft Office software (e.g., Word, Excel, PowerPoint)
Recommended:
- Read articles from academic journals
- Use data analysis software, like R, Python, or STATA
- We are a team, and we value collaboration. If the assistants have any questions, we will discuss and solve the problem(s) together.
- Academic study is rigorous. We should be careful in collecting and analyzing all the data. No fake data.
- We hope that scientific research can also be fun.
{Empty}

UROP Program Elements

Yes
Yes
Yes
Yes
{Empty}
Yes
{Empty}
2023
https://cre.fsu.edu/urop-research-mentor-project-submission-portal?element_parents=elements/research_mentor_information/headshot_optional_&ajax_form=1&_wrapper_format=drupal_ajax&token=a4l_RyRA2xFkn6nuT4n1T7l42zL4Vl17wSKzfbrGWSs