STATS1350WeekOneNotes.pdf Stats 1350
Popular in Statistics 1350
verified elite notetaker
Popular in Department
This 5 page Class Notes was uploaded by Alyssa Leathers on Tuesday January 20, 2015. The Class Notes belongs to Stats 1350 at Ohio State University taught by Ali Miller in Winter2015. Since its upload, it has received 338 views.
Reviews for STATS1350WeekOneNotes.pdf
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 01/20/15
STATS 1350 12015 130 PM Week One Notes Chapter One Individuals objects described by a set of data subjectsparticipants Variables characteristics of individuals different values for different individuals 0 Categorical Variable places an individual into one of several groupscategories words not numbers 0 Quantitative Variables in terms of numbers 0 Ex meet and greet 0 Individuals people 0 Variables categorical name majorquantitative miles fro campus Ways to collect data 0 Observational Study survey passive Experiment impose some treatment ex medical Population set of all individuals that we want info about 0 Census attempts to get info from every individual in a population expensive and time consuming Sample a part of the population to gather data from cheaper quicker easier Parameter mean weight of all dogs in the city Statistic mean weight of a sample of 50 of these dogs Statistical Inference confidence intervals and hypothesis testing Inference the process of drawing conclusions about population from sample Example Meet and greet 0 Population of interest all students in class 0 Sample people you talked to 0 Selecting sample convenience Conclude anything 0 Sample size important Chapter Two Types of Samples 0 Convenience Sample sample individuals who are easily accessible 0 Ex meet and greet 0 Pros Fast cheap easy o Cons Can t make much inference not necessarily representatives 0 Voluntary response sample individuals in the population choose whether or not to be included 0 Ex online opinion poll 0 Pros easy fast cheap ethical medical o Cons most opinionated people are more likely to respond might differ systematically from population bias 0 Random Sample every individual in the population has equal chance of being sampled 0 Ex random name 0 Pros best inference ideal can draw stronger conclusions since sample is more likely to be represented o Cons expensive time consuming unethical 0 Types of Random Sampling 0 SRS simple random sample of size n 1 Need a list ordered of all students population called a sampling frame 2 Select random numbers from the list 3 Choose accordingly 0 Stratified Random Sample 1 Divide population into groups subgroups with something in common strata n Ex gender age etc 2 Do SRS of each groupstrata Useful Accounts for eliminates variability between groups looking ahead smaller margin of error higher power hypothesis test Bigger populations don t necessarily mean bigger samples for good inference Population Size Biased some individuals are more likely to be chosen 0 Sample inference depends on sample size not population size 0 Ex teacher survey over the phone 0 Population all teachers 0 Sample teachers who were calledanswered Statistic the proportion of teachers sampled who self iden ed Parameter proportion of all US teachers who are engaged Stratified sample strata cell phone landline time zone 12015 130 PM 12015 130 PM