Descriptive statistics Flashcards
Draw __ to check distribution of a continuous variable
Histogram
Normal distribution is a probability distribution that is:
symmetric about the center
data near center are more frequent in occurance
a bell curve in graph form
More data on the left tail than right
Skewed to the left
More data on the left tail than right
Skewed to the right
Suggestive of two different populations
Bimodal
Most common value for central tendency
not good for data with an outlier or skewed distribution
Mean
value in the middle of a ranked data
better for data with an outlier or skewed distribution
Median
value that occurs most often
highest bar on histogram
rarely reported for continuous data
Mode
if mode > median > mean
data skewed to right
if mean > median > mode
data skewed to left
Range
maximum - minimum
or as interval (min, max)
common measure for spread of data around median
can be misleader in data with outlier
Interquartile range (IQR)
Q3-Q1 or (Q1, Q3)
where:
Q1= value that occurs at 1st quarter mark
Q2= value that occurs at 2nd quarter mark (median)
Q3= value that occurs at 3rd quarter mark
Measures better than range for data with outlier
If you count the numbers in each category of a table you are measuring:
Frequency
If you find the % of each category of a table you are measuring:
Proportion
Charts for continuous data
Histogram and Box plot