Scribd Logo
Upload

Welcome to Scribd!

  • Data Analysis

    Uploaded by

    baojia chen
    11 views15 pages

    Original Title

    Data analysis

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    Download as pdf
    11 views15 pages

    Data Analysis

    Uploaded by

    baojia chen

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    Download as pdf
    of 15
    fe Probability ‘The probability of an event is the number of ways that the event can occur, divided by the total number of possible ‘outcomes. The symbolic farm is mE) a(S) 1) PCE) represents the probability of event E. P(E)= 2) n(E) represents the number of ways that event & ean occur. 3) n(S) represents the numberof possible outcomes in sample space S. Note: 1) Event is acollection of desirable outcomes ofa procedure 2) Sample space is the set ofall possible cutcomes. 3) Probability isthe likelihood of an event occurring. 4) Probability isalwaysbetween Qand 1. If P=0 —> Impossible event . If P=1 —> Certain event. 5) The complement of event "A" is denoted A —> P(A)+P(A)=1 Example: Procedure Event (desirable outcomes) Sample space (all possible outcomes) Flipa coin onetime Head Fread, Tail Flipa coin three time “Two head and one tail (mp, (ur, ary, THM, GIT, (THT), (TTY, (TTT) ‘Type of Probability: 1) Observed Probability isthe probability that is estimated based on observation. 2) Theoretical Probability is the probability base on the equal chance of each event. ‘Multiplication Rule 1) Independent Events: Iftwa events do not affect each other, the events are independent P(Aand B) = P(4)-P(B) Example: In a deck of $2 cards, what is the probability of selecting a king and a queen, in that order, from the deck with replacement? Solution) With replacement, since the second event is not affected by the firstevent, they are independent. 44 1 P(K and Q)=P(é)-P(Q)=—- = (K and Q) = P(k)-P(Q)= 555 = Tem 2) Dependent events: Ifthe second event is affected by the first event, the events are dependent P(Aand B)=P(4)-P( 5/4), P(5/A) means “Event B is affected by event 4” Example: In adeck of 52 cards, what is the probability of selecting a king and a queen, in that order, from the deck without replacement Solution) Without replacement, two events are dependent. « and O)= Pl AAA P(K and Q)= P(t): P(O/K)= 3-3 = A Example: A bag contains 3 red marbles and 4 blue marbles. What is the probability that youselect one red marble and one bine, at random, from the bag? Solution) ‘There are two different sclections. RBorBR Therefore, 12 4 3_12 R and B) =, P(BandR)=Sx2=* PRandB)= 7x E= 75» PBandR)= > DR 12. 12_ 24 Theansweris + = ans 70 SAT Practice 1. Abagcontains 3 red marbles and3 blue marbles. | 3, In.abox, there are & blue marbles and g green What isthe probability that you draw two red marbles. [f'a person sclects two marbles, what is the marbles without replacement? probability that both marbles are blue? 1 1 1 1 A) B= OZ D- a 6 5 3 ) ie b beget bxb 9 —~ (b+s)(b+8—1) b(b-1) (b+2)(b+2-1) 3 5 9 4. [fa number js chosen at random from the set {-15,-10,—5, 0, 5, 10, 15, 20} , what is the probability that itis a member of the solution set of 2. The three cards shown abave were taken from a box jeezfen? often cards, each with a different integer on it from 1 to 10. What is the probability that the next two cards selected from the box will have bath even integer on (a) 0 iv w+ at gE of ps 4 a 3B 7 7 a2 8 wt Dr John Chungs SAT Math 63Tips 1 Geometric Probability "Geometric probability" is the probability dealing with the areas of regions instead of the "number" of outcomes. The cquatian becomes Prababiiy = Frerable region ‘Area of total region A typical problem might be this: Ifyou are throwing a dart at the rectangular target below and are equally ikely to it any point on the target, what isthe probability that you will hit the small square? 25cm avorabl Total 25cm? 10cm Probability = ‘This means that there isa 1 in 10 chance that a dart thrown at dhe rectangle will hit he small square. SAT Practice 1. Inthe figure above, theradius of thecircle P is2the | 2. Inthe figure above, a circle is inside of and outside of radius of the circle Q is 4,and AB isthe diameter of the largest cirele. Ifa dart is thrown at the circular target and is equally ikely to hit any point on the target, what is the probability that the dart will hit the shaded region? t at 4 square. Ifa point is chosen at random from square ABCD, what is the probability that the point is chosen from the shaded region? A 4 B) B) QO Db) 72 Data Interpreta Data interpretation problems usually requite two basic steps. First, you have to read! a chart or graph in order to obtain certain information. Then, second, you have to apply or manipulate the information in order ta obtain an answer. Be sure to read all notes elated to the data, GEOMETRY TEST RESULTS FOR 10 STUDENTS 100 80 60 Lest 4a} 20] 10) 2 aga ROTO Test 1 In the scatter plot above, student A got 30 om test I and 70 on test 2. Saident # got 70 an test Land 9 on test 2, 1) What is the median score on test 1 and 2? Solution) On test 1 30,40, 50, $0, 60, 70, 80, 80, 80, 100 ‘The median is sor. Solution) On test 2 50, 50, 60, 60, 70, 70, 80,90, 90, 100 ‘The median is 70. 2) What s the average (arithmetic mean) on test 1? Solution) The average is 30440450 +50+60+70-+ 80480480 +100 10 Dr Jonn Ghungss SAT Math 63 Tips. 73 SAT Practice COST VS. WEIGHT FOR 10 MEATS. Cost 1. For 10 meats of different weights, the cost and weight af each are displayed in the seater plot above, and the line af best fit forthe data is shown. Which of the following is closest to the average (arithmetic mean} cast per pound forthe 10 meats? A) S018 B) $056 © C) $0.62 ——D) $0.73 ITEMS PURCHASED BY EACH CUSTOMER Numbers of | Number of Customers | Items 10 10 25 g 4s 5 50. Fewer than 5 2. Theteble above shows the number of items 130 customers purchased froma stationery store during on Sunday. Which of the following can be obtained from the informetion in the table? 1. The average (arithmetic mean) number of items TL. The median number of tems. WL. The mode of the number of items. A) only B) Wonly © Mlonly D) Tand I only 74 le) Box-and-Whisker Plot Box-and-Whisker Plots A box-and-whisker plot displays the extremes, the ‘quartiles, nd the median for a se af data. The length of the bax represents the interquartile range. A vertical line inside thebox represents the median, Horizontal lines (whiskers) representthe lower and upper fourths of the data. The bullets at each endare the extreme values. Example: Draw a box-and-whisker plot for the following data, 3,7,9, 14, 16, 19, 19, 25, 28 Solution) Step 1 Determine the quarties and any outliers, With the data ordered fiom Icast to greatest, determine the «quartiles. The median is 16. The lower quartile, QI, i or 8 ‘The upper quartile, 03, is or 22 “The interquartile range IQR) is 22-8 = 14. (Check to see ifthere are any outliers. An outlier isa data ‘value that isat least 1.5 interquartile ranges less than QI ot Step 2 Drow and label a number line, Use a scale that includes the extreme values, Step 3 Complete the box-and-whisker plot, Draw a box to designate the data between the upper and lower quties Draw aline through the median, Draw a line from the lower ‘quartile to the Ieast value and from the upper quartile to the ercatest value. Draw a box-and-whisker plot for each set of data, Q=M_ O, rr a Sn a aL 0 3 6 9 12 15 18 21 24 27 30 31 Least value = 3 Greatest value= 28 Q=8, Q=M=16,0,=2 rear than 03, JOR =14 8-1,5(14)=13 and 22-41.5(14)=43 ‘There are no numbers less than —13 or greater than 43 in thedata se. Therefore, there arena outliers. SAT Practice 29 30 31 1. For the box-and-whisker plot above, ‘a. What are the extremes of the data? b. What is the range of the data ‘c, What is the median of the data? 4. What is the interquartile range of the data? Dr. John Chung's SAT Math 63 Tips. \ 1 32 33 34 35 36 37 38 39 40 75 Linear Correlation Coefficient ‘The linear correlation coefficient, r, measuresthe strength and the direetion of linear relationship between two variables x and _y for the specific equation of line of best-fit. 1) The correlation coefficient is always a value such that -I<7< 1, 2)Positive comclation: [fvariables x and y- have a srong positive linear comelation, 0.8 strong positive correlation 3)-08<)<08 > weak correlation SAT Practice 1. Inthe scatterplot above, which of the following best represents the correlation coefficient between two variables? A-l B09 G09 D1 2. Whichof the following graphs best represents a strong positive association between x and y? 76 A) B) 9 D) Scatter plot and a Line of Best Fit ‘When data is displayed with a seattemplot, a ine of best fit is useful far purpose of predicting values that ‘may not be displayed on the plot. 1) A line of best fit is straight line that best represents the data on a scatterplot. Ths line may pass through some of the points. 2) The equation ofa line of best fit depends upon the points chasen to construct the fine. 3) Thegraphing caleulatorhas the capability of determining which line will actually representthe line-of: best-fit SAT Practice Questions 1-5 refer to the following informatio) y 650 600 550 500 450 400 350 300 250 200 Calories tee 0-5 10 1520 25 3035 40 45 50 Fat (gram) ‘The scatterplot above shows the lationship between the fat grams and the calories in a fast food. The line of best fit is also shown above. 1. Which of the following best describes the correlation | 3. How many ealories would be predicted for 80 fat seen in the graph? grams? A) High positive correlation Aj 600 B)S40—C) 1000) 1200 B) Low positive correlation ©) High negative correlation D) Low negative correlation 4, Fora 25 fat grams, which of the Following is the percent decrease from the actual calories to the calories predicted by the linc ofbest fit? 2. Whatis the equation ofthe line of best ft? A) 10% B) 15% 20% Dy 25% 5. Which ofthe following is the calories per one fat gram? A) Bealories BY. 10 calories D)y =10x+200 ©) 100 calories D) 200 calories Dr. John Chung's SAT Math 63 Tips 7 Standard Deviation Standard deviation is a measure of dispersion ofa set of data values from the mean, (How far away from mean) Remember: The greater standard deviation, the greater the spread of data values fiom the mean” 1) A standard deviation close t0 0: The data points tend to be very clase tothe mean, 2) A high standard deviation: ‘The data points are spread out aver a wider range of values. 3) Ifall data are the same, standard deviation is zero. Example: (Use your graphing calculator and check each standard deviation) For each data set: {10,10,10,10,10,10,10} > Standard deviation is 0. {8,9, 10, 10, 10,9, 8} > Sample standard deviation .90, Population standard deciation:er, =0.83 (7,8, 9, 10,9, 8 7} Sample standard deviation is 1.11, Population standard deviation is 1.03. Remember: 1) Statistics: Practice or science of collecting and anslyzing numerical data in large quant es, especially forthe purpose of inferring proponions in a population from those representative sample. 2) Data: Any observations that has been collected. 1) Parameter: A characteristic of population ‘Type of data: 2) Statistic: A characteristic of a sample 3) Sample: Some subset ofa population. 4) Population: The complete set af elements being studied. 5) Census: Collecting data fiom every member af a population. 6) Random: Each member af'a population has an equal chance of being selected in the sample, 7) Standard deviation (Normally sample) is a measure of the spread of data, We can estimate the population standard deviation from a sample standard deviation SAT Practice Questions 1 and 2 refer to the following information. DataSet 1: {I, 1, 1,3, 3, 3,3, 5, 6 8} DataSet 2: {2, 2, 3, 3,3, 4, 4, 4, 5, 5} DataSet 3: {3, 3, 3, 3,3, 4, 4, 4, 4, 4} 1. From the data sets above, which data set appears tohave |2. Which data set appears to have the sinallest standard the largest standard deviation? deviation? A) Data set 1 A) Data set 1 B) Data set2 B) Data set2 C) Data set3 ©) Data set3 D) It cannot be determined from the information given. 78 D) Itcannot be determined from the information given. Two-way frequency table ‘Two-way frequeney table represents possible relationships between two sets of categorical data, Gender ‘Male Female Total Going to College 00 4B 108 Not Going to College R 6 18 Total R 54 120 ‘Tombstone high school surveyed 120 senior students about their plans after they graduate. Data are shawn in the two-way frequency table above. Shaded region represents joint frequency (Two variables) and sub-total represents marginal frequency (One variable). Maleand Going to college [ Mand | Fand© | Female and Going to college Malc and Not going to college | -MandN FandN | Female and Not going to college Example: 1) Check the values for the following joint frequencies. (Two variables) n(MandC)=60 n(FandC)=48 = n(MandN)=12 —_n(F and N)=6 2) Check the values forthe following marginal frequencies. (One variable) n(M)=72 n(F)=54 n(C)=108 n(N)=18 3) The probability that a person picked at random would be a female who is going to college. 48 ee insss 4) The probability that a person picked random would be female or someone going to college. n(F or C)_n(F)+n(C)-n(FOC) _54+108-48 _114 P(F or C)= See # Ot Total 120 120 120 5) If person was picked at random, the probability thatthe person was going to college given they are female. n{ F and C] ricinj eG pros —> Conditional probability Note: n(Fand C)sa(FC) and a(F or C)=n(FUC) Dr. Joha Chungs SAT Math €3Tips 79 80 SAT Practice ‘The school board of Smith town in the Southeast, is considering building a new library near the high school primarily finded by local taxes. Every registered voters in the town was surveyed. The data from the survey are summarized in the table, Should Our Town Build a New Library? Yes No No Answer Taal Male 120 80 25 225 Female 186 145, 18 349 Total 306 225 8 57 1. Ifa randomly selected eligible voter is male, what is the | 3. A randomly selected person who completed the survey probability he will vote to build a new library? voted to build a new library. What is the probability this person is female? 120 a 20 ) Sa 1) 186 3 120 374 ” 360 wy 186 306 oF 374 oe 120 225 py 22 ns py 306 5m 2. Tfa randomly selected eligible voter is fernale, what is | 4, What isthe probability thata randomly selected person ‘the probability she will vote to build a new library? voted yes to buildinga new library or was a fernale? 186 3! 349 574 1 186 By) 8 x74 74 306 186 308 o 86 9 574 J 306 306 186, py 06 186 Sao oa iui Kolm Two-way relative frequency table ‘Two-way relative frequency table shows the probability (or percentage) of each of the corresponding events. (The ratio of the actual number of favorable events to the total possible number of events) Gender Male Female Total Goi z ing to College 60 o50 48 g40 168 0.90 120 120 120 Nu GonewG lot Going to College 2 Loto © ans 18 Lois 120 120 120 Teial 2 ogy | 404s 2 120 120 120 1) Check the values forthe following joint relative frequencies. (Two variables) PM OC)=050 P(FC)=040 — P(MandN)=0.10 —-P(Fand N)=0.05 2) Check the values for the Following marginal relative frequencies. (One variable) P(M)=0.60 P(F)=0.45 P(C)=0.90 P(N) =0.15 3) The probability thata person picked at random would bea female wha is going to callege. P(F\C)=0.40 4) The probability that a person picked mndom would he female ar someone going to college, P(F or C)=P(F)+ P(C)—P(F and C)=0.45+0.90-0.40=0.95 5) Ifa person was picked at random, the probability that the person was going to college given they are feral. sm £(FOC)_040_8 < PIC =F = pasta Comltonal prbabiity Note: 1. P(FAC)=P(F and C) and P(FUC)=P(F or C) 2. Conditional probability: The conditional probability afanevent B is the probability thatthe event will acu giventhatanevent A has already occurred. This probebility is written P(2 | 4). Fromthis definition, P(B and 4) P(A) Ifevents A and B are independent, (a ay= eee) 227m) P(B\4)= (Compare with the results in Tip 57.) Dr. John Chung’s SAT Math 63 Tips at SAT Practice ‘A survey was taken to examine the relationship between hair colorand eye color in alittle town, The table shows the relative frequencies of the people surveyed who fell into each category. Tair Calor Black Blond Red Total Blue 0.20 0.18 0.02 040 os Brown 15 1S 0.04 034 os Green 0.15 0.10 0.01 0.26 Total 0.50 0.43 0.07 1 ind the probability hat the person picked had blue | 3. What isthe probability thata person picked at random eyes given they had black bir ‘would have blond hair with brown eyes? 2. Find the probability that the person hod blond hair given they had green eyes. Statistical Study and Sampling Tecknics “Type of statistical study ‘A statistical study bogins by asking a question that can be answered with data, 1, Observational study ‘An observational study is one in which the values of one ‘or more variables are observed with no attemptto affect the outcomes 2. Surveys One kind of observational study is asurvey. A survey requires asking a group of people to respond to one ar more questions. 3. Experiments Suibjcets are assigned to treatments forthe purpose of secing what effect the treatment has on some response. ‘Sampling Technics (Methods) 1. Convenience Sample a. Non-probability sampling method where the sample istaken from a group of people easy to get b. Not random . For example, standing at a mall ara grocery store and asking people to answer questions would be an example of a convenience sample. 4. Convenicnce sampling is not often recommended for research due to the possibility of sampling error ‘and lack of representation of population. 2 Systematic sampling ‘a. Statistical method involving the selection of elements from an ordered sampling fame, b, For example, the sampling stats by selecting an clement from thelist at random and then every k member inthe frame is selected, where the sampling interval where is the sample size, and Nis the population size. 3. Stratified Random Sampling Divide a population into homogeneous subgroups. 4, Cluster sampling Divide a population internally heterogeneous subgroups, Randomly select acertain number of clusters and then collect dhe data from the entire cluster. Sampling Bias 1. Non-random sample of a population. 2 Selection fiom a specific area. 3. For example, ‘A sample is also biased ifcertain members are underrepresented or overrepresented relative to others in the population. SAT Practice 1. Express gas station is conducting a consumer satisfaction survey. Which method of collecting data ‘would most likely lead toa biased sample? A) interviewing every Sth customer to come into the station B) interviewing customers chosen at random by a computer at the checkout ©) interviewing customers who are the VIP members with the gas station, D) interviewing every customer who comes into the station on a day of the week chosen at random, Dr John Chung’s SAT Math 83 Tips 2. In which of the following situations és a survey least likely ta contain bias? A) Surveying a sample of people leaving a theater about their favorite movie B) Surveying a sample of people leaving a elassie concert about their favorit music C) Surveying people leaving a grocery store about their political part affition 1D) Surveying people who use social networking website about theit favorite communication media 83 (Ya Confidence Interval and Margin of Error 1. Population and Sample ‘The “population’ in statistics inchudes all members of a defined group that we are studying or collecting information on for data driven decisions. A part of the population is called a sample Sample means a subgroup of the members of population ‘chosen for participation in the study. 2. Sample mean (7) ‘The arithmetic mean of random sample values drawn from the population is called sample mean, 3. Population mean (1) ‘The arithmetic mean of the entire population ¥ called population mean. 4, Problem between a sample and a population mean, Researcher use inferential statistics ta infer from the sample data what the population might think. But one difficult is thata sample is generally nat identical to the population from which it comes. Specifically, the sample ‘mean will differ from the population mean. 5. Margin of errar(E) Margin of error is the maximum likely difference given a certain confidence level, between the observed sample mean and the truc value of the population mean, 6. Computing the margin of err Margin of error vn Critical 2-value @= Standard deviation Sample size TevelofConfidence | * 0% Las 95% 1.96 9% 2.575 Note: 95% is the most common confidence rate. 7. Therefore, we inféras follows. FE

    You might also like