Transcript Bi-Variate Data
Bi-Variate Data
Honors Analysis
Univariate and Bivariate Data Univariate data: Data of a single data type. (Ex: Test scores, number of wins, age of family members, salaries of employees, etc.) Bivariate data: A set of data comparing pairs of values in an ordered pair. Generally used when a relationship exists between the two data types.
***Does not necessarily imply causality!
Scatter Plots You can show relationships between two variables using a scatter plot.
Correlation – how closely the data follows a pattern
Correlation Coefficient -- r Your graphing calculator can calculate a value r, called the Pearson Product-Moment Correlation. It is a value between -1 and 1 that describes the linearity of a data set.
If r = 1, the data forms a perfect linear relationship with a positive slope. If r = -1, it forms a perfect linear relationship with a negative slope.
1 𝑟 = 𝑛−1 (𝑥 1 −𝑥) (𝑦 1 −𝑦) + 𝑠 𝑥 𝑠 𝑦 (𝑥 2 −𝑥) (𝑦 2 −𝑦) + ⋯ 𝑠 𝑥 𝑠 𝑦
Examples of r values and scatter plots
Writing Linear Equations – Point Slope Form
𝑦 − 𝑦
1
= 𝑚(𝑥 − 𝑥
1 )
Writing Line of Best Fit Equations Least-squares linear regression is a process often used to write best fit equations.
The central concept is trying to find the smallest error (called the residual between each point and the line.
Your graphing calculator allows you to determine linear regression lines using the LinReg function.
Constructing a Confidence Interval for a Sample Proportion Calculate the standard deviation for the sample: 𝑠𝑑 = 𝑛 The standard deviation for the sample proportion is often called the Margin of Error (ME)
Constructing a Confidence Interval for a Sample Proportion Determine the confidence level. You can estimate 68% confidence using ±1𝑀𝐸 , 95% confidence using ±2𝑀𝐸 , and 99% using ±3𝑀𝐸 . For a more specific confidence level, you will need to use your table of z-scores to find create the desired confidence.
𝑍 ∗ , the number of std. dev. above/below the mean to
Constructing a Confidence Interval for a Sample Proportion To calculate a more specific confidence interval, determine scores).
𝑍 ∗ (you may want to sketch a picture of the interval and then use your table of z Confidence Interval: ∗ 𝑛
Calculating Confidence Intervals It is also possible to use the formula for ME to determine the sample size required to attain a desired confidence level.