Descriptive Statistics Uses data to provide descriptions of a population, either through numerical calculations or graphical displays. Sampling › When a population is very large, conducting a census can be difficult, costly, and impractical. › Instead, a subset of members selected from a population, called a sample, is studied How do I select a representative sample i.e. unbiased? › Simple random sampling: every combination of members of a given population has an equal likelihood being chosen › Systematic Sampling: members of the population are put into some order. A starting position is chosen at random and then every kth member in the ordering is selected until a sample size is reached › Stratified Sampling: split the population into strata. Then randomly select the correct number of individuals from each strata so that their proportions when combined together into a single sample would be representative of the population examined › Cluster Sampling: When the population is already organised into clusters that are each representative of the population, randomly selecting some number of these clusters should provide a good sample › Sequence sampling: used frequently in quality control, involves randomly taking a sequence of successive units taken from a production line as the sample Identify the type of sampling used A test to ensure the maximum capacity for a foot chair involves testing "to destruction" every 100th stool from the assembly line (i.e., weight is added to each stool until it collapses) Systematic The "Math:rand" function on the TI-83 calculator generates random decimal values between 0 and 1. A group of 70 people are assigned numbers 00 to 69. A random value is generated on the TI-83 calculator using the aforementioned function. If the first two digits of the random number match someone's assigned number, then that person is selected to be part of a sample. For example, if the number 0.35876397688 was generated, person 35 would be selected to be in the sample. If the first two digits don't match any of the assigned numbers, or if they correspond to someone already selected, nothing happens. This process is repeated until 10 people have been selected to form the entire sample Simple random Numerical Summaries of Data › Well-constructed data summaries and displays are essential to good statistical thinking › It is useful to describe data features numerically › Characterizing the location (or central tendency) and variability (spread or scatter) in the data is an example of a numerical summary › Data are often a sample of observations that have been selected from some larger population of observations – This type of population is called a conceptual or hypothetical population because it does not physically exist Sec 6.1 Numerical Summaries of Data 4 Sample Mean The location or central tendency in the data can be characterized by the arithmetic average or the sample mean. For a finite population with 𝑁𝑁 equally likely values, the probability mass function is 𝑓𝑓(𝑥𝑥𝑖𝑖 ) = 1/𝑁𝑁 and the mean is Sec 6.1 Numerical Summaries of Data 5 Example 6.1 | Sample Mean Consider 8 observations (𝑥𝑥𝑖𝑖 ) of pull-off force from engine connectors as shown in the table. 8 ∑x i = x = average = 8 = 104 = 13.0 pounds 8 i =1 12.6 + 12.9 + ... + 13.1 8 i 1 2 3 4 5 6 7 8 xi 12.6 12.9 13.4 12.3 13.6 13.5 12.6 13.1 13.00 = AVERAGE($B2:$B9) Figure 6.1 The sample mean is the balance point. Sec 6.1 Numerical Summaries of Data 6 Sample Variance and Standard Deviation The variability or scatter in the data may be described by the sample variance or the sample standard deviation. The units of measurement for the sample variance are the square of the original units of the variable, while the standard deviation measures variability in the original units. Sec 6.1 Numerical Summaries of Data 7 Example 6.2 | Sample Variance =VAR.S($B2:$B9) =STDEV.S($B2:$B9) The table displays the quantities needed for calculating the sample variance and sample standard deviation. The numerator of 𝑠𝑠 2 is Sec 6.1 Numerical Summaries of Data 8 Computation of 𝒔𝒔𝟐𝟐 The prior calculation is definitional and tedious. A shortcut is derived here and involves just 2 sums. Sec 6.1 Numerical Summaries of Data 9 Example 6.3 | Shortcut Calculation for For Example 6.2, we calculate the sample variance and standard deviation using the shortcut method. Sec 6.1 Numerical Summaries of Data 10 Exercise 6.1.6 › In Applied Life Data Analysis (Wiley, 1982), Wayne Nelson presents the breakdown time of an insulating fluid between electrodes at 34 kV. The times, in minutes, are as follows: 0.19, 0.78, 0.96, 1.31, 2.78, 3.16, 4.15, 4.67, 4.85, 6.50, 7.35, 8.01, 8.27, 12.06, 31.75, 32.52, 33.91, 36.71, and 72.89. › Calculate the sample mean and sample standard deviation. › Check the values obtained with Excel (available through Blackboard) Exercise 6.1.7 › An article in Human Factors [“Visual Performance on CRT Screens and Hard-Copy Displays” (1989, Vol. 31(3), pp. 247–257)] presented data on visual accommodation (a function of eye movement) when recognizing a speckle pattern on a high-resolution CRT screen. The data are as follows: 36.45, 67.90, 38.77, 42.18, 26.72, 50.77, 39.30, and 49.71. › Calculate the sample mean and sample standard deviation. › Check the values obtained with Excel (available through Blackboard) Exercise 6.1.10 › Data from a second experiment using a low-resolution screen were also reported in the article. They are: 8.85, 35.80, 26.53, 64.63, 9.00, 15.38, 8.14, and 8.24. › Compare it to the one for the first sample. What can you conclude about CRT resolution in this situation? Exercise 6.1.8 › Preventing fatigue crack propagation in aircraft structures is an important element of aircraft safety. An engineering study to investigate fatigue crack in n = 9 cyclically loaded wing boxes reported the following crack lengths (in mm): 2.13, 2.96, 3.02, 1.82, 1.15, 1.37, 2.04, 2.47, 2.60. › Calculate the sample mean and sample standard deviation. › Check the values obtained with Excel (available through Blackboard) Exercise 6.1.11 › The following data are the joint temperatures of the O-rings (°F) for each test firing or actual launch of the Space Shuttle rocket motor (from Presidential Commission on the Space Shuttle Challenger Accident, Vol. 1, pp. 129–131): 84, 49, 61, 40, 83, 67, 45, 66, 70, 69, 80, 58, 68, 60, 67, 72, 73, 70, 57, 63, 70, 78, 52, 67, 53, 67, 75, 61, 70, 81, 76, 79, 75, 76, 58, 31. › Compute the sample mean and sample standard deviation using Excel. › Set aside the smallest observation (31°F) and recompute the quantities. Comment on your findings. How “different” are the other temperatures from this last value? The meaning of 𝒏𝒏 − 𝟏𝟏 in the denominator › The population variance is calculated with 𝑁𝑁, the population size. Why isn’t the sample variance calculated with 𝑛𝑛, the sample size? › The true variance is based on data deviations from the true mean, 𝜇𝜇. › The sample calculation is based on the data deviations from 𝑥𝑥,̅ not 𝜇𝜇. › 𝑥𝑥̅ is an estimator of 𝜇𝜇; close but not the same. › So the 𝑛𝑛 − 1 divisor is used to compensate for the error in the mean estimation. Sec 6.1 Numerical Summaries of Data 16 Degrees of Freedom › When the sample variance is calculated with the quantity 𝑛𝑛 − 1 in the denominator, the quantity 𝑛𝑛 − 1 is called the degrees of freedom › Origin of term: – There are 𝑛𝑛 deviations from the 𝑥𝑥̅ in the sample – The sum of the deviations is zero – 𝑛𝑛 − 1 of the observations can be freely determined but the 𝑛𝑛𝑡𝑡𝑡 observation is fixed to maintain the zero sum Sec 6.1 Numerical Summaries of Data 17 Sample Range strongly affected by extreme values present in the distribution In addition to the sample variance and sample standard deviation, the sample range is a useful measure of variability. For Example 6.3 (pull-off force data), the sample range is 𝑟𝑟 = 13.6 − 12.3 = 1.3. =MAX($B2:$B9) - MIN($B2:$B9) Sec 6.1 Numerical Summaries of Data 18 Group Activity › Compute the range of the previous data sets › Compare the scatter of the range and the standard deviation › What would you conclude about the effect of the measurements on each representation of the variability? Quartiles › The three quartiles partition the data into four equally sized counts or segments. – First or lower quartile : 25% of the data is less than q1. – Second quartile : 50% of the data is less than q2, the median. – Third or upper quartile : 75% of the data is less than q3. › For the Table 6-2 data: Usually not affected by the presence of extreme values in the data set. Denoted as 𝑥𝑥� Descriptive Statistics Mean 162.6625 Minimum Q1 Median, Q2 Q3 Maximum 76 144.5 161.5 181 245 Midrange 84.5 (Sample) Variance 1140.631 (Sample) Standard Deviation 33.77324 Range Mode 158 169 (76 to 245) Interquartile Range (IQR) 36.5 (IQR Test) Lower Outlier (IQR Test) Upper Outlier 89.75 181 SEC 6-2 STEM-AND-LEAF DIAGRAMS 20 Percentiles and Interquartile Range › Percentiles are a special case of the quartiles. › Percentiles partition the data into 100 segments. › The interquartile range (IQR) is defined as: IQR = q3 – q1. › From the Quartiles example: IQR = 181.00 – 143.25 = 37.75 = 37.8 › Impact of outlier data: – IQR is not affected – Range is directly affected. SEC 6-2 STEM-AND-LEAF DIAGRAMS 21 Stem-and-Leaf Diagrams Sec 6.2 Stem-and-Leaf Diagrams 22 Example 6.4a | Alloy Strength › Consider the data in the table. We select as stem values the numbers 7, 8, 9, …, 24. Sec 6.2 Stem-and-Leaf Diagrams 23 Example 6.4b | Alloy Strength › The resulting stem-and-leaf diagram is shown. › Inspection of the diagram reveals that most of the comprehensive strengths lie between 110 and 200 psi and that a central value is somewhere between 150 and 160 psi. › The strengths are distributed approximately symmetrically about the central value Sec 6.2 Stem-and-Leaf Diagrams 24 Group activity › Please attempt one of the following exercises: › 6.2.4 › 6.2.5 › 6.2.9 Frequency Distributions and Histograms › A frequency distribution is a more compact summary of data than a stem – and – leaf diagram › To construct, we must divide the range of the data into intervals, which are usually called class intervals, cells, or bins › Choosing number of bins approximately equal to the square root of the number of observations often works well in practice Considerations: Range = 245 – 76 = 169 Sqrt(80) = 8.9 Trial bin width = 169 / 8.9 = 18.9 Decisions: Number of bins = 9 Bin width = 20 Range of bins = 20 * 9 = 180 Starting point = 70 Sec 6.3 Frequency Distributions and Histograms 26 Frequency Distribution Table Sec 6.3 Frequency Distributions and Histograms 27 Histograms • A histogram is a visual display of the frequency distribution • Provides a visual impression of the shape and distribution of the measurements and information about the central tendency and scatter or dispersion in the data bin frequency • Unequal bin widths will be employed Rectangle height = bin width Sec 6.3 Frequency Distributions and Histograms 28 Histograms When the sample size is large, the histogram can provide a reasonably reliable indicator of the general shape of the distribution or population of measurements from which the sample was drawn the presence of skewness (or outliers) can affect where the measures of the middle are located relative to one another Sec 6.3 Frequency Distributions and Histograms 29 Examples of Distribution Shapes Group activity › Construct a Histogram on Excel using the data sets from exercises: › 6.2.4 › 6.2.5 › 6.2.9 › Adjust the number of bins for a smooth cumulative frequency Box Plots › The box plot is a graphical display that simultaneously describes several important features of a data set, such as center, spread, departure from symmetry, and identification of unusual observations or outliers › Sometimes called box – and – whisker plots › Displays three quartiles › A line, or whisker, extends from each end of the box One might wonder where the 1.5 in the above interval comes from -- Paul Velleman, a statistician at Cornell University, was a student of John Tukey, who invented this test for outliers. He wondered the same thing. When he asked Tukey, "Why 1.5?", Tukey answered, "Because 1 is too small and 2 is too large." Sec 6.4 Box Plots 32 Box Plots Sec 6.4 Box Plots 33 Group activity › Construct a box plot on Excel using the data sets from exercises: – 6.2.4 – 6.2.5 – 6.2.9 › How does the box plot compare in interpretive value to the original stem-and-leaf diagram? › Comment on the possible presence of outliers › Repeat for the data without the extreme outlier and comment. › Compare the distribution of the data with and without the extreme outlier. The most important functions in Excel (Data Summaries) •AVERAGE() •MIN() •MAX() •QUARTILE.INC() •MEDIAN() •VAR.S() [or VAR.P()] •STDEV.S() [or STDEV.P()] •MODE.MULT() Location (central tendency) Variability (spread or scatter) Remarks Mean (arithmetic average) Standard deviation (or variance) Mean represents centre of distribution and usually affected by extreme values and sample size. Likewise, standard deviation is exaggerated by values far from the mean Midrange Range Easy to compute and strongly affected by extreme values Median Inter-quartile range (IQR) Can also be used for ordinal values; usually not affected by extreme values or sample size Mode Number of modes Most frequent data value in the population/sample. Modes can be used even for nominal data. Group Project › It will be about an observational or retrospective study › The assignment aims to collect a set of real data (the number of observations should be more than 40) identify statistical features, and perform relevant analysis assisted by appropriate graphical plots. › Refer to Section 1.2 to reflect on the limitations in your chosen data sets › YOU ARE REQUIRED TO SUBMIT AN EXCEL WORKBOOK › More details on Blackboard Specific Uses of Statistics › Organize, summarize, and present data › Generalize from knowledge of sample data to (probably accurate) knowledge of a larger population › Test hypotheses › Determine relationships between variables › Make predictions from existing data Some Abuses › Quoting statistics based on non-representative samples › Choosing the “middle" value for a sample which most lends itself to your position, when a different “middle" value would be more appropriate › Speaking of changes in a variable in terms of actual values or percentages to either inflate or deflate their importance psychologically. (How happy would you be if your net worth increased by QAR 10 million. What if that only represented a 0.3% increase?) › Using detached statistics like "1/3 fewer carbs" (fewer than what?) › Implying causal connections between variables without a well-designed experiment to back it up (i.e., "Doctors say that taking lipotrim twice a day ''may reduce'' your weight by up to 30 lbs in the first 2 weeks!") › Formatting graphs to mislead the eye › Designing questions to be used on a survey that will bias the results
0
You can add this document to your study collection(s)
Sign in Available only to authorized usersYou can add this document to your saved list
Sign in Available only to authorized users(For complaints, use another form )