Ch 1; Picturing Distributions w/ Graphs Flashcards
Statistics
Science of leaning from data.
Any set of data contains…
Information, organized in variables, about some group of individuals.
Individuals
The objects described by a set of data. (People, animals, things)
Variable
Any characteristic of an individual.
Data
Actual measurements of variables recorded for individuals. (Numbers, pieces of info)
2 Types of Variables
Quantitative, Qualitative/Categorical
Categorical Variables
Places individuals into one of several groups (gender)
Quantitative Variable
Takes numerical values for which arithmetic operations can be used. (Often has units)
2 Types of Categorical Variables & Examples
Ordinal (Size, blood group, performance, pain ratings)(Scaled), Nominal (Colour, gender)
2 Types of Quantitative Variables & Examples
Continuous (2.31, 4.25), Discrete (0,1,2)
Exploratory Data Analysis
Process of using statistical tools & ideas to analyze data in order to describe main features.
Steps to Exploratory Data Analysis
1) Examine Each Variable
2) Examine relationships b/w variables
3) Graph
4) Add numerical summaries of specific aspects of data
Distribution of a Variable
Values it takes & how often it takes these values
What does a Distribution of a Categorical Variable do?
Lists categories & gives either percents/count of individuals in each category. (Values are labels of categories)
Best Graphs for Categorical Variables (2)
Pie Chart, Bar Graphs
Pie Chart
Shows distribution of categorical data as pie where slices are sized by counts/percents for categories of a whole. (Single Whole)
Bar Graph
Each category is a bar whose height shows category counts/% (Single variable/variables from different sources not part of a whole)
Best Graphs for Quantitative Variables (3)
Histograms, Stemplots, Time Plot
Histogram
Show distribution of quantitative variables by using bars whose heights represent #/frequency of individuals who take on a value within a particular class (Range of #s).
Stemplot
Separate each observation into a stem & a leaf that are then plotted to display distribution while maintaining the original values of variable.
What to look for when Describing Distributions
Look for patterns & striking deviations from pattern
How can you Describe Pattern (4)
Shape, Center, Variability, Outliers (Individual values that fall outside of pattern)
Shapes of a Distribution (3)
Symmetric; Right & left mirror images
Right Skewed; Long tail on right
Left Skewed; Long Tail on left
Bimodal; 2 bumps
Time Plot
Shows behavior over time (Time on horizontal axis & measured variable on vertical axis)