1 1 Where Does Data Come From Vocabulary

1. 1 Where Does Data Come From? Vocabulary Review

Statistics Vocabulary • Statistics is the art and science of dealing with data • Anecdotal Evidence is when you base your conclusions on 1 or 2 events that stick out in your mind instead of looking at the data as a whole. It’s bad science. – Source of many stereotypes – “It worked for me, so it must work for anyone”

Population, Sample & Individuals • Population is everyone or everything that you want to study • Sample is the people, animals or things you actually measure and observe • Individuals are the objects (people, animals, things) described by a set of data.

Population, Sample & Individuals • Find 100 women age 30 of which 50 have been smoking a pack a day for 10 years while the other 50 have been smoke free for 10 years. Measure lung capacity for each of the 100 women. • What is the population? – Women age 30 • What is the sample? – 100 women examined • What are the individuals? – Each woman is an individual

Population, Sample & Individuals • A storeowner buys a crate of 100 oranges. He randomly grabs 10 to test if they are good before accepting the shipment. – What’s the population? • The 100 oranges in the crate – What’s the sample? • The 10 oranges the storeowner tested – What are the individuals? • Each orange is an individual

Population, Sample & Individuals • You want gather student opinion on the dress code at Saint Joe. You poll 50 people on their opinion. – What’s the population? • All Saint Joe Students – What’s the sample? • The 50 people you polled – What are the individuals? • Each person you poll is an individual

Surveys & Census • A survey is when you take a sample of a population and ask them questions to learn about the population as a whole • Can a survey be an experiment or is it always an observational experiment? Can you apply a treatment if all your doing is asking questions?

Surveys & Census • A census is a survey where you poll every single individual in the population • The survey you took on the first day, was it a census? – Depends on why I gave the survey. – If I’m using it to gather information about the typical St. Joe student then it is a survey. – If I’m using it to gather information about my class then it is a census

Surveys & Census • A census is a survey where you poll every single individual in the population PROS CONS More Accurate Time Consuming Expensive

Variables • A variable is any characteristic of an individual • What were the measured variables from Wednesday’s survey? – Gender – Height – # Siblings – Favorite Pizza Delivery Company – # Windows – Favorite Music – Right/Left Handed – Time on Internet, Sleeping, Doing Homework – Birth Month

Variables • A categorical variable places an individual into one of several groups or categories • A quantitative variable takes numerical values where arithmetic like averaging makes sense

Variables • Is each Categorical or Quantitative? – Gender Categorical – Height Quantitative – # Siblings Quantitative – Favorite Pizza Delivery Company Categorical – # Windows Quantitative – Favorite Music Categorical – Right/Left Handed Categorical – Time on Internet, Sleeping, Doing Homework Quantitative – Birth Month Categorical

Variables • Find 100 women age 30 of which 50 have been smoking a pack a day for 10 years while the other 50 have been smoke free for 10 years. Measure lung capacity for each of the 100 women. • What were the measured variables and were they quantitative or categorical? – Lung Capacity Quant. - Smoker Y/N Cat. • What variables were held constant and were they quantitative or categorical? Quant. – Age Quant. - Gender Cat. -Length Smoker Status

Observational Study vs. Experiment • Observational Study: is when you observe and measure, but don’t apply a treatment. • Experiment: is where you deliberately impose a treatment to measure a response.

Observation or Experiment? • Find 100 people who have been smoking a pack a day for 10 years • Find 100 people who have never smoked. • Measure lung capacity for the two groups. • Analyze, interpret, and draw conclusions. Observation • You didn’t apply a treatment, just observed. • How would you make this an experiment?

Observation or Experiment? • You want to study the impact of room color on mood. You grab 200 people, and put 100 in a bright red room and 100 in a pale beige room. You then give them a series of psychological questions to judge their mood. Experiment • What is the treatment? • Room Color

Observation or Experiment? • You want to study the impact of question wording? You call 200 people and ask them who they are voting for. The 1 st 100 you ask “Would Romney make a good president” and the next 100 you ask “Is Obama a good president” Experiment • What is the treatment? • Question Wording

Observation or Experiment? • You want to know how income affects political beliefs, so you call 200 people and ask them “Who they are voting for? ” and “What is their income? ”. Observation • You didn’t apply a treatment, just observed. • How could you make this an experiment? • Hire 200 people, assign them a rich or poor income. Then see if it affects how they vote.

Observation vs. Experiment • What are the advantages of doing an experiment? – More accurate – More control = less error • What are the advantages of doing an observational study? – It’s easier – In some cases an experiment would be unethical

Observational Studies Read Example 1. 2 on page 4 -5 • What was the measured variables? – Economic Class and Recycling Weight • What was variables that affect recycling weight? – Recycling Quantity, Recycling Type • How can we redesign this study?

Observational Studies • Read Example 1. 3 on page 6 -7 – Why does it matter that the study occurred over 2 different time periods? – In an experiment you want to be testing only 1 variable at a time. – This experiment has 2 variables: insurance type and time, since the 2 different insurance plans were implemented over 2 different time periods. – How could they redesign this study to eliminate lurking variables?

Observational Study vs. Experiment • In both you only want 1 variable (1 difference) between your 2 groups so you can say with certainty that is the cause of any different results

Observational Study vs. Experiment • In an experiment, the 2 groups ideally start out identical in every way and then you cause the difference • In an observational study, the 2 groups already had a difference without you doing anything and you ideally selected your sample so that ALL other variables are the same

Observation vs. Experiment How would you investigate each of the following relationships? • Caffeine consumption & hours of sleep • Bacon consumption & life expectancy • Parental Income & Adult Education Level • Mood & Clothes Color • Intelligence & Genetics

Homework due Mon. • • • Pg. 5 # 1. 1 -1. 2, 1. 4 Pg. 9 -10 # 1. 8 -1. 10 Pg. 14 -15 # 1. 13 -1. 16, 1. 18 Pg. 17 # 1. 19 -1. 23 Vocab Quiz on Tuesday – Anecdotal Evidence, Observational Study, Experiment, Population, Sample, Individuals, Survey, Census, Variable, Categorical Variable, Quantitative Variable