How does rising atmospheric CO2 affect marine organisms?

Click to locate material archived on our website by topic


Linear Regression Statistics

Equation of the Linear Regression Line
The linear regression line obtained from the statistical output is the "best-fitting" staight line that can be drawn through the data.  It is designated by the equation Y = b1X + b0, where X represents the year, Y represents the predicted temperature anomaly, b1 is the slope of the line and b0 is the Y intercept of the line.

Number of Years
The number of years represents the difference between the starting and ending year over the range of years for which annual temperatures were examined.

Slope (b1)
In the regression equation Y = b1X + b0, the slope (b1) is a constant indicating the magnitude of the change in Y for a unit change in X.  If the slope is equal to one, it can be stated that a unit of change in X will produce a unit of change in Y.  If the slope is greater than one, a unit of change in X will result in more than a unit change in Y.  For a slope less than one, a unit of change in X will result in less than a unit change in Y.  If the slope is zero, a unit change in X will produce no change in Y.

Intercept (b0)
The intercept is the Y value obtained when X = 0.

Explained Variance, R2 (%)
The explained variance indicates the percent of the total variation in temperature that is accounted for by time.  For example, if the R2 is 50%, then time accounts for half of the variation in temperature during the selected period.

F-value and Degrees of Freedom
These two statistics are used to evaluate the statistical significance of the regression model.  Click here for a table that will help determine whether or not this regression equation is significant.