STATS1350WeekSixNotes.pdf Stats 1350
Popular in Statistics 1350
verified elite notetaker
Popular in Department
This 6 page Class Notes was uploaded by Alyssa Leathers on Thursday February 19, 2015. The Class Notes belongs to Stats 1350 at Ohio State University taught by Ali Miller in Winter2015. Since its upload, it has received 98 views.
Reviews for STATS1350WeekSixNotes.pdf
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 02/19/15
STATS 1350 21915 120 PM Week Six Summary Statistics Chapter 12Review Exam 15 TrueFalse and 3 short answer covers basically all materials learned so far HWlabbooklecture examplesetc More on Standard Deviation 0 Where do you think the center is on the histogram o 25 0 Larger standard deviation data far from center 0 Small distance from center most is close to center low standard deviation 0 Standard deviation average distance from the mean how far away are the data from the middle point 0 Symmetric meanmedian roughly the same 0 Which histogram would have the larger standard deviation or would they be the same 0 Range big to small 0 Larger standard deviation more spread out from center What do you think 0 Imagine two classes A and B Each has 11 students 0 Quiz 1 A mean of 75 Standard deviation 203 n Median 814 n IQR 3 B mean of 75 standard deviation 203 n Median 704 n IQR 271 0 Summary statistics are not enough Matching Histograms to Summary Statistics 6 histograms with 6 different sets of information Match up 0 1c2e3a4b5f6d Symmetric similaralmost same mean and median 0 Measure of center Skew right mean larger than median Skew left median larger than mean Matching Histograms to Boxplots A2 B3 C4 D1 0 Median should be around bump in histogram Graph Examples 0 Numerical Data skewhistogramstem pot never pie chartbar graph 0 Category of order bar graph 0 Distribution of favorite hobbies 0 Distribution of women s heights Relatively distributive Choosing Appropriate Measures 0 Mean or Median be higher 0 Mean goes with tail 0 Salaries of 100 employees and 2 executives right skewmean gt median 0 Ages of death infant to elderly left skewmedian lt mean 0 Prices of new cars in one month right skewmean gt median o Heights of seven year ods symmetricalmean median 0 Shoe sizes of adult women symmetrical Something to think about 0 Dependent on type of information Example Broccoli eaters get more out of lifequot 0 Population All adults 0 Sample 405 college students Withinabout sample statistic Withinabout population parameter 0 Observational Study Review HW Standard deviation can never be negative Ex 0 Some individuals did not respond to the survey nonsampling error nonresponse error 0 Response error measurement that isn t correct 0 Asian Americans were underrepresented in the sample sampling error under coverage Ex Sample of 3021 adults6 are aware of recommendations 0 Margin of error 1square root of 3021 addsubtract that from 6 o 95 confidence statement We are 95 confident that the true proposition of all adults who are aware of the recommendations is between and Ex 0 A biased measuring process is also considered unreliable FALSE 0 Biased tends to overunder estimate 0 Reliable all measurements are similar low random error 0 If random error is small in a measuring process we say it is reliable vs unbiased RELIABLE Ex 0 Days attending church is not an appropriate way to measure commitment to churchquot days attending churchquot is An INVALID MEASURE of religious commitment Ex 5 number summary 0 2 5 11 14 19 19 25 31 36 45 75 Boxplot scale first Distribution Right skew right arm longer median is to the left of the box 0 Outliers o IQR Q3 Q1 3611 0 Q115 x IQR 1115 x25 low outlier o Q315 x IQR 3615x25 upper outlier 0 Median element 6 1112 19 0 Q1 element 3 512 11 0 Q3 element 3 512 36 0 Min 2 0 Max 75 O O Ex Ex Individuals college students Explanatory variable type of program watched Response level of aggression Experiment Applying treatments to the individuals controlling the explanatory variable Random assignment Avoid lurking variablescofounding Concluding correlation no observational studies can t control for lurking variables Cofounding when effects on a response can t distinguished Explanatory variable type of light including no light Lurking variable genetics 21915 120 PM 21915 120 PM