The framing of initial COVID‐19 communication: Using unsupervised machine learning on press releases

• Published date: 01 September 2023
• Author: Stella Tomasi,Sushma Kumble,Pratiti Diddi,Neeraj Parolia
• Date: 01 September 2023
• DOI: http://doi.org/10.1111/basr.12323

ORIGINAL ARTICLE

The framing of initial COVID-19

communication: Using unsupervised machine

learning on press releases

Stella Tomasi

1

| Sushma Kumble

2

| Pratiti Diddi

3

|

Neeraj Parolia

1

1

Department of Business Analytics and

Technology Management, Towson

University, Towson, Maryland, USA

2

Department of Mass Communication,

Towson University, Towson, Maryland,

USA

3

Department of Communication, George

Mason University, Fairfax, Virginia, USA

Correspondence

Stella Tomasi, Department of Business

Analytics and Technology

Management, Towson University,

Towson, MD, USA.

Email: stomasi@towson.edu

Funding information

Towson University

Abstract

The COVID-19 pandemic was a global health crisis that

required US residents to understand the phenomenon,

interpret the cues, and make sense within their envi-

ronment. Therefore, how the communication of

COVID-19 was framed to stakeholders during the early

stages of the pandemic became important to guide

them through specific actions in their state and subse-

quently with the sensemaking process. The present

study examines which frames were emphasized in the

states' press releases on policies and other COVID

information to influence stakeholders on what to focus

on to help with sensemaking during the crisis. We con-

ducted content analysis on 602 press releases from

50 US states using an unsupervised machine learning

approach called Latent Dirichlet Allocation (LDA). The

results show that health communication using press

releases to help the public make sense of the crisis were

framed to include health frames as well as economic

frames. Health communication messages are typically

framed with health and safety measures; however, this

study shows that economic frames were emphasized

more than public health frames in the government's

health communication for COVID-19, which forced

both large and small businesses to engage in specific

Received: 13 November 2022 Accepted: 30 June 2023

DOI: 10.1111/basr.12323

© 2023 Albert P. Viragh Institute for Ethics in Business at Duquesne University.

Bus Soc Rev. 2023;128:515–531. wileyonlinelibrary.com/journal/basr 515

socially responsible activities that were previously vol-

untary to support public health safety.

KEYWORDS

COVID-19, framing, government communication, machine

learning, press releases, sensemaking, social responsibility

1|INTRODUCTION

The rapid spread of the novel COVID-19 virus around the world created a major pandemic. The

novel virus shifted all aspects of the normal flow of daily life for all individuals across the world.

The world has adopted various mitigation strategies like social distancing, self-isolation, COVID-

19 testing, travel restrictions, closure of businesses, the introduction of mask mandates, and other

measures to help reduce the number of cases and deaths. When individuals are confronted with

unexpected situations, like the novel COVID-19, they strive to understand what is happening and

interpret the cues within their environment (Maitlis & Christianson, 2014). The American

response to the unexpected virus, COVID-19, was marked by the division of power between the

US state government and the federal government (Bergquist et al., 2020). Each state was leading

the efforts on containment and enforcing health policies while helping individuals and businesses.

Neelon and his colleagues found that the differences in COVID-19 infection, death, and testing

varied across the states based on political affiliations (Neelon et al., 2021) and not by public health

guidelines. By March 2020, all states in the United States had de clared a state of emergency, which

allowed governors to use emergency powers that included initiating emergency workforces and

funds to improve healthcare access (Bergquist et al., 2020). It was also seen that instead of the fed-

eral government outlining the policies, states were given the ability to determine the content of

policies and the time of their implementation (Bergquist et al., 2020; Neelon et al., 2021). There-

fore, US governors were the initial actors and leaders making important decisions during the early

stages of the pandemic (Voges & Binford, 2021). While current research on COVID-19 focused on

varied policies during the state emergency (for example, Béland et al., 2021), our study instead

looks at the communication and messaging that were provided during the onset of the COVID-19

crisis when there was limited information provided by the federal government and little was

known about the virus. For example, by 2020, the Center for Disease Control (CDC) stated that

wearing a mask was an effective way to slow the spread of COVID-19 (CDC, 2020); however, Sur-

geon General Jerome Adams said that wearing a mask can increase a person's chance of getting

COVID-19 (Schreckinger, 2020). Therefore, it was important to see how states were communicat-

ing with their stakeholders during the early stages of the pandemic.

In the following paragraphs, we discuss how states used frames for their messages to help the

public with sensemaking of their new environmentand policies. The CDC developed a communi-

cation model called the Crisis and Emergency Risk Communication (CERC) model in 2001. The

model identified the best communication practices during the different stages of a health crisis.

Initially, people are unsure how the crisis will unfold and how it will affect their lives. Typical cri-

sis situations are identified with fear, panic, and anxiety (Kayes, 2004). Therefore, the model sug-

gests that efforts need to be made to educate the public with messages that allow them to make

sense of the crisis and frame messages to reduce uncertainty around the issue (Veil et al., 2008).

While the messaging may change as more information about the crisis is discovered, it is

516 TOMASI ET AL.