Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Chapter 6 The Standard Deviation and The Normal Model How can we compare? 1. A sister is 67 inches tall. Her brother is 72 inches tall. Who is taller (comparatively)? 2. You took SAT and scored 1410. Your friend took ACT and scored 30. Which score is better? Solution – compare by standardizing those values, Bring them to “common denominator”! Slide 6- 2 The Standard Deviation as a Ruler: The trick in comparing very different-looking values is to use standard deviations as our rulers The standard deviation tells how the whole collection of values varies, therefore it is a natural ruler for comparing an individual to a group Slide 6- 3 Standardizing with z-scores We compare individual data values to their mean, relative to their standard deviation using the following formula: y y z s We call the resulting values standardized values, denoted as z. They can also be called z-scores. Slide 6- 4 Standardizing with z-scores (cont.) IMPORTANT: Standardized values have no units. z-scores measure the distance of each data value from the mean in standard deviations. A negative z-score tells us that the data value is below the mean, while a positive z-score tells us that the data value is above the mean. Slide 6- 5 When Standardizing, Standardized values have been converted from their original units to the standard statistical unit of standard deviations from the mean. Thus, we can compare values that are measured on different scales, with different units, or from different populations. Slide 6- 6 Question1: A sister is 67 inches tall. Her brother is 72 inches tall. Who is taller (comparatively)? Height of women y 66, s y 2.5 Height of men x 70, sx 3 Slide 6- 7 Standardizing Variables y y 67 66 z 0.4 sy 2.5 x x 72 70 z 0.67 sx 3 A sister is 0.4 standard deviations above mean height for women. Her brother is 0.67 standard deviations above mean height for men. The brother is taller (comparatively). Slide 6- 8 Question 2: SAT vs. ACT You took SAT and scored 1410. Your friend took ACT and scored 30. Which score is better? SAT has mean 1150 and standard deviation 100. ACT has mean 18 and standard deviation 6. Slide 6- 9 Standardizing: SAT vs. ACT Your score Friend’s score 1410 1150 2.6 100 30 18 2 6 You scored better on SAT than your friend did on ACT. Slide 6- 10 What is the standardizing actually? - it is just Shifting and Rescaling Data Shifting data: Adding (or subtracting) a constant amount to each value just adds (or subtracts) the same constant to (from) the mean. This is true for the median and other measures of position too. In general, adding a constant to every data value adds the same constant to measures of center and percentiles, but leaves measures of spread unchanged. Slide 6- 11 Shifting Data (cont.) The following histograms show a shift from men’s actual weights to kilograms above recommended weight: Slide 6- 12 Rescaling data: When we divide or multiply all the data values by any constant value, both measures of location (e.g., mean and median) and measures of spread (e.g., range, IQR, standard deviation) are divided and multiplied by the same value. Slide 6- 13 Rescaling Data (cont.) The men’s weight data set measured weights in kilograms. If we want to think about these weights in pounds, we would rescale the data: Slide 6- 14 Back to z-scores y y z s Standardizing data into z-scores shifts the data by subtracting the mean and rescales the values by dividing by their standard deviation. Standardizing into z-scores does not change the shape of the distribution. Standardizing into z-scores changes the center by making the mean 0. Standardizing into z-scores changes the spread by making the standard deviation 1. Slide 6- 15 When Is a z-score Big? A z-score gives us an indication of how unusual a value is because it tells us how far it is from the mean. The larger a z-score is (negative or positive), the more unusual it is. Slide 6- 16 When Is a z-score Big? (cont.) There is no universal standard for z-scores, but there is a model that shows up over and over in Statistics. This model is called the Normal model (You may have heard of “bell-shaped curves.”). Normal models are appropriate for distributions whose shapes are unimodal and roughly symmetric. Slide 6- 17 Normal Model We write N(μ,σ) to represent a Normal model with a mean of μ and a standard deviation of σ. We use Greek letters because this mean and standard deviation do not come from data - they are numbers (called parameters) that specify the model. Summaries of data, like the sample mean and standard deviation, are written with Latin letters. Such summaries of data are called statistics. When we standardize Normal data, we still call the standardized value a z-score, and we write z y Slide 6- 18 Normal Model Once we have standardized, we need only one model: The N(0,1) model is called the standard Normal model (or the standard Normal distribution). Don’t use a Normal model for just any data set check the following condition: Nearly Normal Condition: The shape of the data’s distribution is unimodal and symmetric. This condition can be checked with a histogram or a Normal probability plot.** Slide 6- 19 The 68-95-99.7 Rule Normal models give us an idea of how extreme a value is by telling us how likely it is to find one that far from the mean. In a Normal model: about 68% of the values fall within one standard deviation of the mean; about 95% of the values fall within two standard deviations of the mean; and, about 99.7% (almost all!) of the values fall within three standard deviations of the mean. Slide 6- 20 The First Three Rules for Working with Normal Models Make a picture. Make a picture. Make a picture. And, when we have data, make a histogram to check the Nearly Normal Condition to make sure we can use the Normal model to model the distribution. Slide 6- 21 Example: pg. 125 #18 Some IQ tests are standardized to a Normal model, with a mean of 100 and a standard deviation of 16. a. b. c. d. e. Draw the model for these IQ scores. Clearly label it, showing what the 68-95-99.7 Rule predicts about the scores. In what interval would you expect the central 95% of the IQ scores to be found? Answer: 68 to 132 IQ points About what percent of people should have IQ scores above 116? Answer: 16% About what percent of people should have IQ scores between 68 and 84? Answer: 13.5% About what percent of people should have IQ scores Answer: 2.5% above 132? Slide 6- 22 Finding Normal Percentiles by Hand – see textbook*** When a data value doesn’t fall exactly 1, 2, or 3 standard deviations from the mean, we can look it up in a table of Normal percentiles – review the textbook. Slide 6- 23 Finding Normal Percentiles Using Technology Most calculators and statistics programs have the ability to find normal percentiles for us. Look under: 2nd DISTR You will see THREE “norm” functions that know the Normal model. They are: normalpdf( normalcdf( invNorm( Slide 6- 24 normalpdf( Calculates y values for graphing a Normal curve. You won’t use this often, if at all. Try graphing: Y1=normalpdf(x) Use graphing window xmin = -4, xmax = 4, ymin = -.1, ymax = .5 Slide 6- 25 normalcdf( Finds the area between two z-score cut points, by specifying normalcdf(zleft,zright) You will use this function OFTEN!! Slide 6- 26 Example: Using normalcdf Find the area of the shaded region between z = -0.5 and z = 1.0. (Always draw a picture first.) -.5 Calculator entry: Lower: Upper: 1 μ: 0 σ: 1 Paste normalcdf(-.5,1.0), or normalcdf(-.5,1.0,0,1) = 0.532807 Slide 6- 27 Example: SAT Scores Again Question 2: SAT II scores are described by model N(500,100). What proportion of SAT II scores fall between 450 and 600? 1. Make a picture of this normal model. Shade desired region. 2. Find z-scores for the cut points (-0.5 and 1). 3. Use your calculator to find this area. 4. The other method!!! – normalcdf(450,600,500,100)=0.532807, 53.3% Slide 6- 28 To infinity and beyond! (not really) Theoretically, the standard Normal model extends right and left forever. Since you can’t tell the calculator to use infinity as a right of left cut point, the book suggests that you use 99 (or -99). What area of the Normal model does normalcdf(.67,99) represent? What area of the Normal model does normalcdf(-99.1.38) represent? Slide 6- 29 Example: pg. 127 #34 – IQs revisited (Cont.) Based on the Normal model N(100,16) describing IQ scores, what percent of people’s IQs would you expect to be: Answer: 89.4% a. Over 80? b. Under 90? Answer: 26.6% c. Between 112 and 132? Answer: 20.4% Slide 6- 30 From Percentiles to z-Scores in Reverse Sometimes we start with areas and need to find the corresponding z-score or even the original data value. Example: What z-score represents the first quartile in a Normal model? Slide 6- 31 invNorm( Finds the z scores of a specific percentile, in decimal form Find the z score at the 25th percentile (first quartile): invNorm(.25) = -.6744897495 The calculator says that the cut point for the leftmost 25% of a Normal model is approximately z = -0.674. Slide 6- 32 Example: SAT once again Question 3: Suppose a college says it admits only people with SAT II verbal test scores among the top 10%. How high a score does it take to be eligible? 1. Draw a picture and shade appropriate region. 2. What percentile will we be using? 3. Use calculator to find z score: invNorm(0.9) 4. Convert z score back to the original units, OR…. invNorm(0.9,500,100) Answer: z = 1.28, and the score is 628 points on the SAT. Slide 6- 33 How Can You Tell - Normal or not? When you actually have your own data, you must check to see whether a Normal model is reasonable. Looking at a histogram of the data is a good way to check that the underlying distribution is roughly unimodal and symmetric. A more specialized graphical display that can help you decide whether a Normal model is appropriate is the Normal probability plot: If the distribution of the data is roughly Normal, the Normal probability plot approximates a diagonal straight line. Deviations from a straight line indicate that the distribution is not Normal. Slide 6- 34 How Can You Tell - Normal or not? (cont.) Nearly Normal data have a histogram and a Normal probability plot that look somewhat like this example: Slide 6- 35 How Can You Tell - Normal or not? (cont.) A skewed distribution might have a histogram and Normal probability plot like this: Slide 6- 36 What Can Go Wrong? Don’t use a Normal model when the distribution is not unimodal and symmetric. Don’t use the mean and standard deviation when outliers are present—the mean and standard deviation can both be distorted by outliers. Don’t round off too soon too much. Don’t round your results in the middle of a calculation. Don’t worry about minor differences in results. Slide 6- 37 What have we learned? The analysis of the data can be easier after shifting or rescaling the data. Shifting data by adding or subtracting the same amount from each value affects measures of center and position but not measures of spread. Rescaling data by multiplying or dividing every value by a constant changes all the summary statistics—center, position, and spread. The power of standardizing data: Standardizing uses the SD as a ruler to measure distance from the mean (z-scores). With z-scores, we can compare values from different distributions or values based on different units. z-scores can identify unusual values among data. Slide 6- 38 What have we learned? (cont.) Before using a Normal model - check the Nearly Normal Condition with a histogram or Normal probability plot. Normal models follow the 68-95-99.7 Rule, and we can use technology or tables for a more detailed analysis. Slide 6- 39