Widgetized Section

Go to Admin » Appearance » Widgets » and move Gabfire Widget: Social into that MastheadOverlay zone

Using Machine Learning to Evaluate Public Policy

The views expressed are those of the author and do not necessarily reflect the views of ASPA as an organization.

By April Heyward
April 15, 2021

Public policy is multifaceted and can be characterized as the vehicle to administer the actions of government to the public. Policy affects all organizational sectors and individual life sectors. Environmental policy is an intervention to address environmental issues. It affects relations between organized interests and the government as well as other policy areas such as health policy and economic policy. Education policy is an intervention to address economic inequality and social inequality. It affects the construct of Pre-K, K-12 and higher education in public institutions and private institutions. Welfare policy is an intervention to improve the quality of life and to provide assistance in response to needs. Public policy serves as a nod to the Social Contract Theory put forward by John Locke in his book, Two Treatises of Government. Social contract entailed individuals giving up some of their rights and freedoms to the government, who in turn act on behalf of the will of the people. Locke postulated that the government should serve the people. Machine learning is a method for evaluating public policy and the public opinion of public policy.

Public policy can be employed to address wicked problems such as the COVID-19 pandemic. COVID-19 was classified as a pandemic in March 2020 due to the exponential growth in disease transmissions, cases and deaths globally. There were varying response times and reactions which contributed to the state of COVID-19. Numerous complexities influenced policy response approaches to include science versus politics, economy versus public health, media versus public health, conspiracy theories, mixed signals and infrastructure challenges. The driving force of policy construct are ideas and the actions required to employ ideas. There are a plethora of approaches to policy development (e.g., instruments), implementation and changes. Policy instruments are public policy vehicles and are characterized as the procedures employed by the government to administer interventions to the public. Anneliese Dodds (2018) described how policy instruments can facilitate the provision of resources, authority and organization and the flow of information in her book, Comparative Public Policy.

The United States employed several public policy interventions to combat the COVID-19 pandemic and its impact. Policy responses can be segmented into the federal government level and the state and local government level. The focus is the federal government level. The interventions employed by the federal government include, but are not limited to, the CARES (Coronavirus Aid, Relief, and Economic Security) Act, PPPHCEA (Paycheck Protection Program and Health Care Enhancement) Act, CPRSA (Coronavirus Preparedness and Response Supplemental Appropriations) Act, FFCRA (Families First Coronavirus Response) Act and the American Rescue Plan Act of 2021. The interventions are a combination of health, economic, educational and social welfare policies. The American Rescue Plan is the latest intervention and builds upon the policy response to the impact of COVID-19. Some of the features of the American Rescue Plan include COVID-19 vaccinations; economic relief to small businesses, state and local governments and individuals; unemployment benefits, housing assistance and agriculture and nutrition programs. One of the most popular features of the American Rescue Plan is the stimulus payments of up to $1,400.00 to individuals.

April Heyward employed machine learning as a method to answer the question of how data science and machine learning informs public policy about COVID-19 policy responses. Machine learning is a fascinating subdiscipline of artificial intelligence and its vastness facilitates natural language processing to execute sentiment analysis. Heyward has been examining the public opinion of COVID-19 policy interventions over several months by extracting Twitter data into RStudio, which is an Integrated Development Environment for R (programming language). Twitter data on the American Rescue Plan was extracted on February 16, 2021, March 10, 2021 and March 18, 2021. Retweets were excluded from extraction. The data was cleaned by removing punctuations, http links, stopwords, digits, other symbols and capital letters which were transformed to lowercase letters. Sentiments were extracted from the Tweets and ten emotions were plotted. See Figure 1 for American Rescue Plan Tweets Emotions Analysis 2-16-2021, Figure 2 for American Rescue Plan Tweets Emotions Analysis 3-10-2021 and Figure 3 for American Rescue Plan Tweets Emotions Analysis 3-18-2021.

Figure 1 – American Rescue Plan Tweets Emotions Analysis – 2-16-2021

Figure 2 – American Rescue Plan Tweets Emotions Analysis – 3-10-2021

Figure 3 – American Rescue Plan Tweets Emotions Analysis – 3-18-2021

It is important to note the timeline of the American Rescue Plan to extend comprehension of the emotions analyses. The American Rescue Plan was proposed on January 14, 2021 and the U.S. House started working on the budget resolution on February 2, 2021. The bill was formally introduced in the U.S. House on February 24, 2021 and passed the U.S. House on February 27, 2021. The U.S. Senate passed the bill on March 6, 2021 but sent the amended bill back to the U.S. House, which formally agreed to the amended bill on March 10, 2021. The American Recue Plan was inked into law on March 11, 2021 and the IRS started processing stimulus payments on March 12, 2021. Anticipation and positive emotions were the top emotions in the analyses as responses to this intervention. Trust, surprise, joy, negativity, fear, sadness, disgust and anger were the highest when the U.S. House agreed with the amended bill and later sent it forward for signature. Machine learning is a method for public policy to embrace and employ.


Author: April Heyward is a 3rd Year Doctor of Public Administration Student at Valdosta State University and a Public Sector Practitioner in South Carolina. She can be reached at [email protected] and followed on Twitter: https://twitter.com/heyward_april. For more information on April, visit www.aprilheyward.com. All opinions and views are her own and does not reflect the views and opinions of her affiliations.

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading...

Leave a Reply

Your email address will not be published. Required fields are marked *