BUSA343 - Exploring Data

Key Concepts and Summary Notes

Population: The entire group of entities or observations of interest. Example populations include all potential voters or all households in a study.
Sample: A representative subset of the population, often chosen randomly.
Variable: A characteristic or measurement of a member of a population, such as age, income, or survey responses.
Observation: A complete set of variable measurements for a single individual or entity.
Data Set: A structured collection of data, typically in a rectangular array format.

Numerical Data: Continuous or discrete measurements that can be subjected to arithmetic operations. Examples include salary, height, or age.
Categorical Data: Non-numeric categories or groups. Categorical variables can be nominal (no natural order) or ordinal (with a natural order).

Measures of Central Tendency: Include mean (average), median (middle value), and mode (most frequent value).
Measures of Variability: Include range (difference between max and min), interquartile range (IQR), variance, and standard deviation.
Percentiles and Quartiles: Percentiles indicate the value below which a percentage of observations fall. Quartiles divide data into four equal parts.

Histograms: Used to display the frequency distribution of numerical data.
Box Plots: Summary statistics indicating distribution, median, quartiles, and potential outliers. Useful for comparative analysis.
Scatterplots: Graphs that illustrate relationships between two numerical variables, indicating correlation.

Correlation: A measure of the strength and direction of the linear relationship between two numerical variables. Correlation values range from -1 (perfect negative correlation) to +1 (perfect positive correlation).
Covariance: Similar to correlation but sensitive to the scale of measurement.

Pivot Tables: Useful for breaking down data by different categories, providing summary statistics like averages, counts, or sums. They allow for dynamic analysis of large data sets, easily enabling filtering and sorting based on categorical data.

Analyzed U.S. Presidential elections from 2000-2012, investigating how popular votes translated into electoral victories.
Key data included vote margins, states won by each candidate, and analysis of wins in swing states.
Findings revealed that while candidate A may win the popular vote, candidate B can win the electoral vote due to state distribution and voting rules.

Illustrated variables such as age, gender, income, and opinions regarding environmental policy.
Utilized crosstabs to analyze relationships between categorical variables, reinforcing how age may influence opinions or behaviors.

Outliers: Extreme values that can skew results; need careful consideration when analyzing data sets.
Missing Values: Affect statistical analysis; common methods include ignoring missing values or substituting with means.

Mastery of descriptive statistics, correlation, and data representation is crucial in effectively analyzing data and making informed conclusions. Tools like pivot tables enhance the ability to gather insights from extensive data sources.