Linear regression (continued) In a test-retest situation, almost always the bottom group on the 1 st test will on average show some improvement on the.

Slides:



Advertisements
Similar presentations
Stat 1301 More on Regression. Outline of Lecture 1. Regression Effect and Regression Fallacy 2. Regression Line as Least Squares Line 3. Extrapolation.
Advertisements

Linear Regression By Hand Please view this tutorial and answer the follow-up questions on loose leaf to turn in to your teacher.
A 4-6 What is Regression and Median Fit
The Standard Normal Curve Revisited. Can you place where you are on a normal distribution at certain percentiles? 50 th percentile? Z = 0 84 th percentile?
Stat 13 lecture 23 correlation and regression A cartoon from handout The taller the father, the taller the son Tall father’s son is taller than short farther’s.
Normal distribution. An example from class HEIGHTS OF MOTHERS CLASS LIMITS(in.)FREQUENCY
STAT 100 Section Week 1 and 2. SAMPLE QUESTION #1 Suppose we measure the amount of weight 5 Harvard Football players can bench-press, and we record the.
Chapter 6: Standard Scores and the Normal Curve
Math 3680 Lecture #19 Correlation and Regression.
Correlations and scatterplots -- Optical illusion ? -- Finding the marginal distributions in the scatterplots (shoe size vs. hours of TV) Regressions --
MA-250 Probability and Statistics
LSRL Least Squares Regression Line
Optical illusion ? Correlation ( r or R or  ) -- One-number summary of the strength of a relationship -- How to recognize -- How to compute Regressions.
LINEAR REGRESSIONS: Cricket example About lines Line as a model:
LINEAR REGRESSIONS: About lines Line as a model: Understanding the slope Predicted values Residuals How to pick a line? Least squares criterion “Point.
Copyright ©2005 Brooks/Cole, a division of Thomson Learning, Inc. Bell-Shaped Curves and Other Shapes Chapter 8.
Chapter 11: Random Sampling and Sampling Distributions
Chapters 10 and 11: Using Regression to Predict Math 1680.
Jan 21 Statistic for the day: The width of train tracks is 4 feet 8.5 inches. Why? Assignment: Read Chapter 9 Exercises from Chapter 8: 16, 18 These slides.
Chapters 8 and 9: Correlations Between Data Sets Math 1680.
Section 2.2, Part 1 Standard Normal Calculations AP Statistics Berkley High School/CASA.
Chapter 6 The Standard Deviation as a Ruler and the Normal Model.
Continuous Distributions. The distributions that we have looked at so far have involved DISCRETE Data The distributions that we have looked at so far.
Density Curves Can be created by smoothing histograms ALWAYS on or above the horizontal axis Has an area of exactly one underneath it Describes the proportion.
1 Everyday is a new beginning in life. Every moment is a time for self vigilance.
Pumpkin Days Day 1 Name:___________________________Math Hr:_______ Science Hr.:_______ Pumpkin # __________ Type of pumpkin: ___________ Directions: 1.Measure.
Measures of Variability Percentile Rank. Comparison of averages is not enough. Consider a class with the following marks 80%, 80%, 80%, 90%, 20%, 70%,
Relationships If we are doing a study which involves more than one variable, how can we tell if there is a relationship between two (or more) of the.
Find out where you can find rand and randInt in your calculator. Write down the keystrokes.
IE(DS)1 Many of the measures that are of interest in psychology are distributed in the following manner: 1) the majority of scores are near the mean 2)
3.2 - Least- Squares Regression. Where else have we seen “residuals?” Sx = data point - mean (observed - predicted) z-scores = observed - expected * note.
STA291 Statistical Methods Lecture LINEar Association o r measures “closeness” of data to the “best” line. What line is that? And best in what terms.
The Normal Distribution Section 8.2. The Galton Board Developed in the late 19 th century by Sir Francis Galton, a cousin of Charles Darwin Theorized.
Stats 95. Normal Distributions Normal Distribution & Probability Events that will fall in the shape of a Normal distribution: –Measures of weight, height,
Chapter 3: Central Tendency. Central Tendency In general terms, central tendency is a statistical measure that determines a single value that accurately.
Example: set E #1 p. 175 average ht. = 70 inchesSD = 3 inches average wt. = 162 lbs.SD = 30 lbs. r = 0.47 a)If ht. = 73 inches, predict wt. b)If wt. =
Copyright ©2005 Brooks/Cole, a division of Thomson Learning, Inc. Bell-Shaped Curves and Other Shapes Chapter 8.
The Standard Normal Distribution Section Starter Weights of adult male Norwegian Elkhounds are N(42, 2) pounds. What weight would represent the.
©2011 Brooks/Cole, Cengage Learning Elementary Statistics: Looking at the Big Picture1 Lecture 35: Chapter 13, Section 2 Two Quantitative Variables Interval.
Chapter 4 Lesson 4.4a Numerical Methods for Describing Data
Modeling Distributions
3.5 Applying the Normal Distribution – Z Scores Example 1 Determine the number of standard deviations above or below the mean each piece of data is. (This.
Karl W Broman Department of Biostatistics Johns Hopkins Bloomberg School of Public Health What is regression?
The Normal Distribution Chapter 2 Continuous Random Variable A continuous random variable: –Represented by a function/graph. –Area under the curve represents.
STANDARDIZED VALUES: LESSON 21. HOW CAN YOU USE STANDARDIZED VALUES TO COMPARE VALUES FROM TWO DIFFERENT NORMAL DISTRIBUTIONS? STANDARDIZED VALUE TELLS.
Normal Distribution A common way that measured data is distributed. It is measured with the MEAN and STANDARD DEVIATION Symmetrical graph The Relative.
Chapter 3 Modeling Distributions of Data Page 101.
Chapter 131 Normal Distributions. Chapter 132 Thought Question 2 What does it mean if a person’s SAT score falls at the 20th percentile for all people.
Write linear equations from tables by identifying the rate of change and the initial value.
Inference for Regression
z-Scores, the Normal Curve, & Standard Error of the Mean
Normal Distribution.
LSRL Least Squares Regression Line
Chapter 4 Correlation.
z-Scores, the Normal Curve, & Standard Error of the Mean
Stat 1301 Percentiles and the Normal Distribution
Regression Fallacy.
Direction: Students who take longer to run the sprint typically have shorter jumps. This means there is a negative association between sprint time and.
Many of the measures that are of interest
Simple Linear Regression
The Normal Distribution
Does Height affect Shoe Size?
Section 2.2 Standard Normal Calculations
Chapter 5 LSRL.
Normal Curve 10.3 Z-scores.
The Normal Distribution
The Normal Distribution
What is regression?.
7.2 Solving Word Problems with Equations
Presentation transcript:

Linear regression (continued) In a test-retest situation, almost always the bottom group on the 1 st test will on average show some improvement on the 2 nd test. The top group will on average fall back. This is known as the regression effect. The regression effect pulls values closer to the average. So if they were above average on the 1 st test, they will be pulled down closer to the average on the 2 nd test.

When we make a scatterplot of these situations it will be “football” shaped.

Using the height of fathers and sons example from the book, we see that for larger x values, there are more points below the SD line. Therefore, on average taller fathers have shorter sons. For the shorter fathers, more points are above the SD line. Therefore, on average, shorter fathers have taller sons.

Examples: Ch. 10 set D #1-3 p. 174 We can compensate for the regression effect by using linear regression to make predictions.

Using the data from the fathers versus sons situation: average ht. of fathers = 68 inchesSD = 2.7” average ht. of sons = 69 inchesSD = 2.7” r = 0.5 How tall would the son of a 62 inch father be?

Example: set E #1 p. 175 average ht. = 70 inchesSD = 3 inches average wt. = 162 lbs.SD = 30 lbs. r = 0.47 a)If ht. = 73 inches, predict wt. b)If wt. = 176 lbs., predict ht. c)Suppose we know the 80 th percentile height. What is the 80 th percentile weight?