COVID 19 Mask and Stay at Home Mandate Project

OVERVIEW

Analysis of mask mandate and stay at home recommendation data to determine whether mandates are effective in mitigating COVID in the United States.

OBJECTIVE

Determine impact of stay at home and mask mandates during the early stages of the pandemic, before vaccines.

ASSUMPTIONS

States with higher numbers of days with mask and stay at home mandates will have lower cases of COVID.

DATA

  • US Centers for Disease Control and Prevention

SKILLS

  • Creating New Variables
  • Integrating Complex and Large Datasets
  • Data Cleaning, Transformation, and Integration
  • Statistical Hypothesis Testing
  • Visual Storytelling

TOOLS

  • Python
  • Tableau

CHALLENGES

The United States confirmed its first case of the 2019 novel coronavirus – or COVID-19 – on January 21, 2020 and the first death on February 29, 2020. This was the early stages of what would go on to become a major global pandemic. Several counties in the US introduced mask and stay at home mandates to prevent the spread of COVID-19. This project examined the impact of these mandates on COVID cases. The analysis starts on April 24 because reliable county-level data for all states is available from the CDC starting on that date. It ends December 13 because the first vaccine was administered in the US the following day, and this project seeks to determine effectiveness of mandates alone. During this time, over 15 million people contracted COVID and nearly 250,00 people died of COVID.

ANALYSIS

One-tailed hypothesis testing determined that states with a higher number of days of with mask and stay at home mandates saw lower cases of COVID, the relationship was weak – in the case of mask mandates, and not statistically significant in the case of stay at home mandates. The correlation between mask mandates and COVID cases per 100,000 people is 0.42 and between stay at home mandates and COVID cases per 100,000 people is 0.30. These relationships are weak, so the null hyothesis cannot be rejected.

It is important to understand the impact of COVID-19 during the research period. Examining cases per 100,000 allows comparisons between states since populations vary by state. A choropleth map analysis illustrates the impact across the US.

Analysis of the frequency of mandates showed that on most days during the research period, staying at home was not required, only recommended. States with no stay at home orders at all were the second most common. Further, 53% of the time, masks were not required in public, while 47% of the time masks were required.

Regarding facemark mandates, analysis indicates that initially, when there were more US counties had facemask requirements, COVID cases were lower. In November, however, cases rose regardless of the mandates. Statistical analysis backs this up – the correlation between mask mandates and COVID cases as 0.42, so there was a weak relationship between COVID cases and mask mandates.

For stay at home mandates, days where staying at home was not required or only recommended saw far higher numbers of cases per 100,000 people, but again the relationship was not statistically significant because the correlation was 0.30.

RECOMMENDATIONS

While scientific research strongly supports that keeping social distance and wearing a mask prevents the spread of COVID-19, mask and stay at home mandates did not have a statistically significant relationship on the number of COVID cases in the US in the early stages of the virus before vaccines.

COVID-19 has an incubation period of 5 to 10 days. Initial studies during the pandemic suggested that counties with mandates saw effects 15, 30, and 60 days later. Further exploration that considers the incubation period may find a stronger relationship between mandates and a reduction of cases.