Stats tests Flashcards
(8 cards)
What are the pros of Chi-Squared Test?
Tests for association between categorical variables.
Good for frequency data in tables.
Simple once the expected values are clear.
Cons of Chi-squared?
Needs large sample size for reliability.
Can’t use with small expected frequencies (<5).
Only for categorical data, not numerical.
Pros of interquartile ranges?
Resistant to outliers (unlike range).
Focuses on the middle 50% — shows spread clearly.
Good for skewed data
Negatives of interquartile ranges?
Ignores extreme values (could miss key variation).
Not as informative alone — best with median.
Harder to calculate by hand than range.
Pros Standard deviation?
Uses all data points — gives a detailed view of spread.
Good for comparing variability between datasets.
Pairs well with the mean in symmetrical distributions.
=> Small SD = Clustering towards towards mean
Cons of Standard deviation?
Sensitive to outliers — can distort results.
Assumes normal distribution (not always realistic).
More complex to calculate than range or IQR.
-More efficient to compare to other
-Extreme values (km^2 etc..)
What is chi squared used for?
Checking if there’s a significant association between two categorical variables (e.g., rock type and land use).
Expected versus observed
What is standard deviation used for?
Measuring how spread out values are around the mean in a dataset.