scatter-plot, best-fit line, and correlation coefficient
TRANSCRIPT
![Page 1: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/1.jpg)
Scatter-plot, Best-Fit Line, and
Correlation Coefficient
![Page 2: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/2.jpg)
Definitions:• Scatter Diagrams (Scatter Plots) – a graph that
shows the relationship between two quantitative variables.
• Explanatory Variable – predictor variable; plotted to the horizontal axis (x-axis).
• Response Variable – a value explained by the explanatory variable; plotted on the vertical axis (y-axis).
![Page 3: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/3.jpg)
Why might we want to see a Scatter Plot?
• Statisticians and quality control technicians gather data to determine correlations (relationships) between two events (variables).
• Scatter plots will often show at a glance whether a relationship exists between two sets of data.
• It will be easy to predict a value based on a graph if there is a relationship present.
![Page 4: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/4.jpg)
Types of Correlations:• Strong Positive Correlation – the values go up from
left to right and are linear.• Weak Positive Correlation - the values go up from
left to right and appear to be linear.• Strong Negative Correlation – the values go down
from left to right and are linear.• Weak Negative Correlation - the values go down
from left to right and appear to be linear.• No Correlation – no evidence of a line at all.
![Page 5: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/5.jpg)
Examples of each Plot:
![Page 6: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/6.jpg)
How to create a Scatter Plot:• We will be relying on our TI – 83 Graphing
Calculator for this unit!• 1st, get Diagnostics ON, 2nd catalog.• Enter the data in the calculator lists. Place the
data in L1 and L2. [STAT, #1Edit, type values in]• 2nd Y= button; StatPlot – turn ON; 1st type is
scatterplot.• Choose ZOOM #9 ZoomStat.
![Page 7: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/7.jpg)
Let’s try one:SANDWICH Total Fat (g) Total Calories
Grilled Chicken 5 300
Hamburger 9 260
Cheeseburger 13 320
Quarter Pounder 21 420
Quarter Pounder with Cheese 30 530
Big Mac 31 560Arch Sandwich Special 31 550
Arch Special with Bacon 34 590
Crispy Chicken 25 500
Fish Fillet 28 560
Grilled Chicken with Cheese 20 440
![Page 8: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/8.jpg)
The Correlation Coefficient:
• The Correlation Coefficient (r) is measure of the strength of the linear relationship.
• The values are always between -1 and 1.• If r = +/- 1 it is a perfect relationship.• The closer r is to +/- 1, the stronger the
evidence of a relationship.
![Page 9: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/9.jpg)
The Correlation Coefficient:
• If r is close to zero, there is little or no evidence of a relationship.
• If the correlation coef. is over .90, it is considered very strong.
• Thus all Correlation Coefficients will be:-1< x < 1
![Page 10: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/10.jpg)
Salary with a Bachelors and AgeAge Salary (in thousands)22 $ 3125 $ 3528 $ 29.528 $ 3631 $ 4835 $ 5239 $ 7845 $ 55.549 $ 6455 $ 85
![Page 11: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/11.jpg)
Find the Equation and Correlation Coefficient
• Place data into L1 and L2• Hit STAT• Over to CALC.• 4:Linreg(ax+b)• Is there a High or Low, Positive or Negative
correlation?
![Page 12: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/12.jpg)
Movie Cost V.Gross (millions)TITLE $ COST U.S. GROSS1. Titanic (1997) $200 $600.82. Waterworld (1995) $175 $88.253. Armageddon (1998) $140 $201.64. Lethal Weapon 4 (1998) $140 $129.75. Godzilla (1998) $125 $1366. Dante's Peak (1997) $116 $67.17. Star Wars I: Phantom Menace
(1999)$110 $431
8. Batman and Robin (1997) $110 $1079. Speed 2 (1997) $110 $4810. Tomorrow Never Dies (1997) $110 $125.3
![Page 13: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/13.jpg)
Finding the Line of Best Fit:
• STAT → CALC #4 LinReg(ax+b)• Include the parameters L1, L2, Y1 directly after it.– (Y1 comes from VARS → YVARS, #Function, Y1)
• Hit ENTER; the equation of the Best Fit comes up. Simply hit GRAPH to see it with the scatter.
![Page 14: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/14.jpg)
Using the Best-Fit Line to Predict.
Once your line of “Best fit” is drawn on the calculator, it can be used to predict other values.
On the TI-83/84:1) 2nd Calc2) 1:Value3) x= place in value
![Page 15: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/15.jpg)
Hypothesis Testing:
• Is there evidence that there is a relationship between the variables?
• To test this we will do a TWO-TAILED t-test
• Using Table 5 for the level of Significance, and d.f. = n – 2; degrees of freedom.
• Compare the answer from the following formula to determine if you will REJECT a particular correlation.
21 2
nr
rt
![Page 16: Scatter-plot, Best-Fit Line, and Correlation Coefficient](https://reader031.vdocument.in/reader031/viewer/2022032203/56649dd35503460f94ac9d05/html5/thumbnails/16.jpg)
TI-83/84 HELP
TI Regression ModelsRules for a ModelDiagnostics OnCorrelation CoefficientCorrelation Not CausationResiduals and Least Squares
Graphing ResidualsLinear RegressionLinear Regression w/ Bio DataExponential RegressionLogarithmic RegressionPower Regression