1 / 15

Stat 301 – Statistics 1

Stat 301 – Statistics 1. Day 2: p-values. Recap – Technology issues. Studio usage rules PolyLearn page Initial course survey. Recap – Syllabus Issues. Calendar Office hours Textbook Glossary links Investigation solutions. Last Time .

helmut
Download Presentation

Stat 301 – Statistics 1

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Stat 301 – Statistics 1 Day 2: p-values

  2. Recap – Technology issues • Studio usage rules • PolyLearn page • Initial course survey

  3. Recap – Syllabus Issues • Calendar • Office hours • Textbook • Glossary links • Investigation solutions

  4. Last Time • Study looked at the choices made by 16 infants, 14 picked the “helper” toy • Possible explanations • Something to the theory that infants have a more positive reaction to the social interaction • Something about the color, size, position of the helper toy – but this was “balanced” out in the study design • Random chance/coincidence/fluke

  5. Last Time • To help us decide whether this could believably be a fluke outcome, we need to know what fluke outcomes look like • How many infants pick the helper when they are just picking randomly between the two? • What is the typical “chance variation” in those results? • “Null model”

  6. Last Time • We can evaluate the “fluke” explanation by simulating the study using the model that infants were picking at random equally between the two toys (50/50) • Know will center at 8, but how rare is 14? • Our coin flip results (16 tosses each) indicated that 14 heads was a bit unusual  Strong evidence to rule out the “fluke” explanation

  7. Improving the simulation • But we only did this a few times. • How can we get a better estimate of how often 14 heads happens “just by chance”? • Practice Problem 1 • The probability of an event is the long-run proportion of times the event will happen if the same random process is repeated infinitely man times

  8. Random Babies applet

  9. Improving the simulation • So how do we run thousands of repetitions of the coin tossing model? • One Proportion Inference Applet • Textbook applets page or Course Schedule page • Set Number of tosses to 16 and press Toss Coin • Uncheck “animate” box and press Toss Coins 4 more times • Answer questions (q) and (s) – 1000 repetitions

  10. One Proportion Inference applet • This distribution by itself tells us nothing! Key is that it tells us how to evaluate the extremeness of the observed result. • One way to measure how extreme an observation is to calculate how often you get results at least as extreme as the one observed. • p-value

  11. Evaluating the p-value • p-value conveys the strength of evidence against the null model • Small p-values are evidence against the claim about the world (just by chance) • Statistically significant • How small? (p. 15)

  12. But can we get an exact p-value? • Binomial probability • Using technology • Minitab • R • RStudio

  13. Investigation 1.2 • Data collection • Answer questions (a)-(c)

  14. Quick Survey • You will be shown two pictures. • Decide which face belongs to “Tim” and which to “Bob”

  15. To do for Wednesday • Practice problem 1.1 (p. 17) • In PolyLearn, under Assignments • Read Terminology Detour (p. 15) and Summary (p. 17) • Review online solutions to Inv 1.1 • Be working on HW 1 (due Thursday/Friday)

More Related