Doing Data Analysis

Intro to Data Analytics

Data Analysis

• Key steps:
  • Define/refine the question
  • Explore the data
  • Develop (formal/statistical) models
  • Interpret results
  • Communicate results

Six types of research questions

• Descriptive: summarize a characteristic of a set of data

• Exploratory: analyze to see if there are patterns, trends, or relationships between variables (hypothesis generating)

• Inferential: analyze patterns, trends, or relationships in representative data from a population

• Predictive: make predictions for individuals or groups of individuals

• Causal: whether changing one factor will change another factor, on average, in a population

• Mechanistic: explore “how” as opposed to whether

Six types of research questions

• Descriptive - summarize characteristics

• Exploratory - analyze patterns

• Inferential - infer properties (w/statistical tests)

• Predictive - make predictions

• Causal - does a change b?

• Mechanistic - how does a change b?

Ex: COVID-19 and Vitamin D

• Descriptive: frequency of hospitalizations due to COVID-19 in a set of data collected from a group of individuals

• Exploratory: examine relationships between a range of dietary factors and COVID-19 hospitalizations

• Inferential: examine whether any relationship between taking Vitamin D supplements and COVID-19 hospitalizations found in the sample hold for the population at large

• Predictive: what types of people will take Vitamin D supplements during the next year

• Causal: whether people with COVID-19 who were randomly assigned to take Vitamin D supplements or those who were not are hospitalized

• Mechanistic: how increased vitamin D intake leads to a reduction in the number of viral illnesses

Starting a data analysis

• Do you have the data to answer the research questions?
• Are there confounding variables?
• Was there any bias in the data collection?

Exploratory data analysis steps

• Generate question
• Read in data
• Check data dimensions
• Head/tail of data
• Plot (a bunch)
• Validate with outside data
• Iterate

Communicating for your audience

• Avoid:
  • Jargon, uninterpreted results, lengthy output
• Pay attention to:
  • Organization, presentation, flow
• Don’t forget about:
  • Code style, coding best practices, meaningful commits
• Be open to:
  • Suggestions, feedback, taking (calculated) risks