What is the purpose of exploratory data analysis (EDA)?
Exploratory data analysis examines a dataset’s structure, summarizes important features, and uses graphs to reveal patterns and questions for further investigation.
Study 6 Exploratory Data Analysis with 12 free online flashcards. Review key terms, definitions, and concepts with this interactive flashcard deck.
What is the purpose of exploratory data analysis (EDA)?
Exploratory data analysis examines a dataset’s structure, summarizes important features, and uses graphs to reveal patterns and questions for further investigation.
What should a focused data-analysis question specify?
Specify the outcome, possible explanatory variables, population and time period, and the groups or conditions to compare.
Why check what a numeric-coded variable represents?
A numeric code may represent a category rather than a measured quantity, so interpreting its numerical summaries as measurements can be misleading.
How should you handle an unusual observation?
First determine whether it is an error or a genuine, informative case; deleting it automatically may remove useful information.
Why might the median better represent a skewed distribution than the mean?
The mean can be pulled toward unusually large or small values, while the median is often more representative for a skewed distribution.
How can you summarize a categorical variable?
Count observations in each category; proportions are useful when comparing groups.
What can histograms and dot plots reveal about numerical data?
Histograms and dot plots display the shape of numerical distributions, helping reveal features such as skew, clusters, and gaps.
What features of a numerical distribution does a box plot show?
A box plot compactly shows the median, quartiles, and potential extreme observations.
What display is useful for examining two numerical variables?
Use a scatterplot to examine their association, including its direction, shape, strength, clusters, and unusual points.
How can you compare a numerical outcome across categorical groups?
Use side-by-side box plots, dot plots, or distributions by group to compare typical values, spreads, and overlap.
How can you explore the relationship between two categorical variables?
Use a contingency table or grouped bar chart to compare category proportions across groups.
What patterns can a time plot reveal?
A time plot can reveal trends, seasonality, abrupt changes, or gaps in measurements over time.