Weekend HW!

If you were absent today you missed a MC classwork that you'll have to make up Monday or Tuesday during lunch, study, or after school.

This weekend = more t practice!

On Monday we'll talk a little more about the t-distribution, then we'll discuss how we handle our conditions if the sample size is not "large enough." And then....that's it! We'll move on to our next chapter

Weekend HW:

• Please complete AND GRADE the "Credit Cards" slide you got in class (or it's below)
• The answer key (to grade) is at the bottom of this post
• You can glue this in your notes and add it there, or do this on separate paper--whatever you think is best for you!
• I think it would be a good idea to do questions 1/2 side by side for this context,
• Show everything! Do each step of the interval/test process
• But you only need to check the conditions once!

Want to get ahead? Here's Monday's HW:

And finally, here's the answer key you need to grade your weekend HW:

Thursday HW

Tonight's (Thursday) HW will count as our first HW grade of the 4th quarter--so it counts double!

But first, a few reminders to wrap up the 3rd quarter:

• If you were out today please remember to bring your stamps tomorrow (or you'll earn a 0, and this is a biiiiig classwork grade)
• If you did not have the hw due today (crows problem and scored) you can show me tomorrow for full credit (check the previous blog post) 
• If you find any more stamps tonight bring those tomorrow too!
• If you have any tests (like Tuesday's) or quizzes to make up they must be done by 2:10 tomorrow

Thursday HW: Please complete and score the 2008 (Form B) Free Response provided in class (or below).

• 1/2 credit = complete the FR
• 1/2 credit = Score
• Use the answer key posted below (rather than the AP rubric) to grade your work
• Part (a) = "new stuff" (kinda)
• One "observation" is one piece of data--so the total number of "observations" we can afford is the sample size we can afford
• There are two ways to approach this...
• One way is to find the necessary sample size (n) to achieve the given a margin of error, then decide if we can afford that sample size...
• We'll use the same conceptual approach as finding n with proportions, but the formula for standard deviation is different (no more p's....)
• We can't find t* without sample size, so you'll just use z* for your critical value in this case
• Be sure to write a sentence that answers the question!
• Part (b) = think about your answer to (a)!
• Part (c): "statistical support' and "statistical evidence" imply a hypothesis test!
• Hypotheses, Conditions, Math, Interpret here....

Today's Class Recap:

• Stamp = Error Analysis for a t-test, t-interval
• Stamps were collected for the 3rd quarter
• HW Questions?
• What changes with our interval formula (with a t-interval)?
• The T-Distribution (Notes):
• What is the t-distribution? Why do we need it?
• How do we calculate degrees of freedom?
• How does df affect the shape of a t-distribution?
• When do we use a t test/interval for means?
• When do we use a z test/interval for means?
• Tomorrow's Focus: (top 2, middle left slides)
• Finding n with means (questions about 2008 part a + stamp)
• How do we find critical values for the t-distribution?
• How do we calculate p-values for the t-distribution?
• How do critical values/p-values compare for t distributions and the Normal model?
• What if we want to do a t-test/interval but the sample size is less than 20 and we're given the data set?

Here's Thursday's HW:

And here's the key to grade yourself after you complete the question:

HW Due Thursday

If you were absent today (Tuesday) you must make up your test before the end of the day Friday or you will have a 0 (quarter ends Friday). 

You have two responsibilities for Thursday....

1.) STAMPS ARE DUE THURSDAY!

• This is our largest classwork grade for the entire quarter, so get these in! Don't forget!
• Get all of your stamps "fastened" together with a paperclip, staple, put them in a folder, an envelope, a bag, or cut them all out and glue them to one paper
• Count your stamp total (twice) and write your name with the total on the front
• If you do not write the total on the front you will be given a score of "1." It's your job to count them!
• I will randomly select 5 names (with the random name generator) for each class and double check those counts!
• If your count is off you will be penalized -15 stamps. Seems harsh? Count better. :)

2.) Please complete AND GRADE the "Intro to T-Tests and T-Intervals (for Means)" assignment we started (in class) on Monday!

• This is our last homework assignment of the quarter--so I'm counting it double!
• The answer key is at the bottom of this post
• You will only earn half credit if you do not grade your answers (use another color so it's clear--"check" anything that's correct, fix anything that's incorrect)
• If you were out the assignment is below
• We did take a tiny bit of "intro notes" on Monday--those are also below (copy those down before tackling the assignment)

On Thursday we'll start discussing the t-distribution: see you there!

Here are the intro (to Unit 6) notes from Monday:

And here is the hw assignment (started Monday) with the answer key:

TEST TOMORROW!

1.) STUDY! TEST TOMORROW!

• Test includes...
• Matching Vocab
• AP FR
• AP MC
• The AP MC and FR will definitely have some "2 prop z" stuff, but can also include any topics from our last test!
• Here's the vocab list for tomorrow (chapters 19, 20, and 21)
• We did not have a chapter 21 vocab quiz, so some of these terms have not been on passed quizzes
• Confidence Interval
• Hypothesis Test
• Sample Statistic v. Population Parameter (given a value determine if it's a statistic or parameter; be sure you also know the symbols for each statistic or parameter)
• Sampling Variability
• Margin of Error
• Critical Value
• Standard error
• Interpret C. Level v. Interpret C. Interval
• Point Estimate
• Know how changing our confidence level and/or our sample size affects the ME and the width of an interval
• Null Hypothesis
• Alternative Hypothesis
• One v. Two Tailed Test
• Alpha Level/Significance Level
• Reject Ho (when do we reject? there is or is not evidence?)
• Fail to Reject Ho (when do we fail to reject? there is or is not evidence?)
• P-Value
• Standardized Test Statistic
• Type I Error
• Type II Error
• Alpha (in relation to types of error)
• Beta (in relation to types of error)
• Know how alpha and beta relate
• How can we decrease both alpha and beta?
• Power 
• Know when to use a 1 prop z test, 1 prop z interval, 2 prop z test, or 2 prop z interval
• There will be another matching section where you have to match the correct procedure to a given context
• Other stuff to know....I'm typing this as I look at the (not vocab section of the) test....
• How to do a full 2 proportion z interval (conditions, math, interpret)
• How can we use a 2 prop z interval to determine if there is a significant difference in two proportions?
• How to conduct a full 2 proportion z test (hypotheses, conditions, math, interpret)
• Describe a Type 1 and Type II error (for 2 prop z contexts)
• Calculate margin of error (for 1 or 2 prop z)
• How to find sample size (back to 1 prop z stuff)
• Be sure you know how to find p-hat and the margin of error given only a confidence interval
• Study the weekend homework! Key in the post below!
• And be sure to look over the 3 multiple choice (stamps) I passed back today! 

In period D we discussed the calculation of the standardized test statistics (z or t score)--check out these notes!

WEEKEND HW!

This weekend we have one more round of practice with our "2 prop z" stuff before we start our next chapter on Monday.

1.) This weekend please complete the "Inference for Two Proportions" worksheet provided in class (or at the bottom of this post).

• For the randomization condition just state that "we'll assume these are random samples of low/normal birth weight children."
• The answer key for this assignment is also provided at the bottom of this post--always check your answers!

2.) Please brainstorm (and write down) any lingering questions you have about one or two proportion z-intervals and tests--on Monday I can answer any questions before we move on to the new stuff!

3.) STUDY! TEST TUESDAY!

• Test includes...
• Matching Vocab
• AP FR
• AP MC
• The AP MC and FR will definitely have some "2 prop z" stuff, but can also include any topics from our last test!
• Here's the vocab list for Tuesday (chapters 19, 20, and 21)
• We did not have a chapter 21 vocab quiz, so some of these terms have not been on passed quizzes
• Confidence Interval
• Hypothesis Test
• Sample Statistic v. Population Parameter (given a value determine if it's a statistic or parameter)
• Sampling Variability
• 10% Condition (how and why?)
• Success/Failure Condition (how and why?)
• Randomization Condition (how and why?)
• Margin of Error
• Critical Value
• Standard error
• Interpret C. Level v. Interpret C. Interval
• Point Estimate
• Know how changing our confidence level and/or our sample size affects the ME and the width of an interval
• Null Hypothesis
• Alternative Hypothesis
• One v. Two Tailed Test
• Alpha Level/Significance Level
• Reject Ho (when do we reject? there is or is not evidence?)
• Fail to Reject Ho (when do we fail to reject? there is or is not evidence?)
• P-Value
• Standardized Test Statistic
• Type I Error
• Type II Error
• Alpha (in relation to types of error)
• Beta (in relation to types of error)
• Know how alpha and beta relate
• How can we decrease both alpha and beta?
• Power 
• Statistically Significant
• A result is statistically significant if we reject the null and determine that there is evidence of the Ha
• Know when to use a 1 prop z test, 1 prop z interval, 2 prop z test, or 2 prop z interval
• There will be another matching section where you have to match the correct procedure to a given context

Weekend HW + Key:

Thursday HW!

Tonight you have two tasks for homework:

HW Task #1: Finish the "side by side" hypothesis test and confidence interval for the "CC v. CC+MMR" context we started today in class.

• If you were out this context is below or it is our last slide for this chapter

HW Task #2: Complete the (5) multiple choice questions provided in class or below. (titled "Chapter 22 MC: Inference for Differences in Proportions")

• SHOW WORK FOR QUESTIONS 3 AND 5; I may collect this tomorrow and grade it (or I may not); but if I collect/grade you will not earn credit for 3 or 5 without the correct work
• For the 3rd question (bottom of front page) it's all about vocab--here we have to know what the "standard error" is...(this came up today in class)
• For the top question on the back (#4) we need to find the pvalue from our calculator; this will narrow it down to 2 options. From there, we have to know the definition of p-value to choose the correct answer
• And for the last question we need the p-value to determine what our conclusion would be; however, we are only given the "test statistic" (z score)....
• So, we have to find the p-value only given the z score...we did a packet once about finding pvalues only given z scores (this was also the bonus on our test)
• We also have to consider the alternative hypothesis to decide if this is a one or two tailed test, which will affect our pvalue
• Once we get the pvalue we can decide if we would reject or fail to reject the null, and then we can decide if there "is or is not evidence..."

Another Stamp Opportunity (DUE TOMORROW):

• You should have two leftover slides from our last chapter...one is about testing for ESP at Duke University and the other is about therapeutic touch (TT).
• Stamp: Write the hypotheses for each scenario. Write the hypotheses on the slide (front or back).
• Both hypotheses should be ultimately written using mathematical notation/numbers, but you could start by considering the hypotheses in words...
• For example, with the Duke University study we're trying to see/prove if someone has ESP--so our alternative is "has ESP," and we assume (in our null) that a person does not have ESP.....
• And now somehow we have to translate these ideas into mathematical notation about "proportions of times the person identifies the card correctly...."
• If you lost the slides you cannot earn these stamps. 

Today's Class Recap:

• Stamp = Review/Practice with our snoring context....
• Interpret C. Level
• Find ME
• How can we increase/decrease ME?
• Interpret Power
• Check out these answers--you can use these as generic templates for interpreting power and confidence level:
• 
• HW Questions?
• Ch. 22 Classwork:
• Used the context on our last slide to do a hypothesis test and a confidence interval for two proportions, side by side (this is what we're finishing for homework)
• Tomorrow in class we'll do one more round of practice, answer any lingering questions we have, and then it's on to our next chapter (23) for Monday

Remember--we have another test (our last of the quarter) coming on Tuesday! 

• Matching Vocab
• AP MC (anything related to tests and intervals, but definitely more emphasis on the 2 prop z stuff)
• AP FR(anything related to tests and intervals, but definitely more emphasis on the 2 prop z stuff)

HW Task #1: Finish the "side by side" hypothesis test and confidence interval for the "CC v. CC+MMR" context we started today in class.

HW Task #2: Complete the (5) multiple choice questions provided in class or below. (titled "Chapter 22 MC: Inference for Differences in Proportions")

Wed Nes Day Hw

Tonight = more practice like last night....2 THOROUGHLY DONE problems in the textbook...

Tonight's HW:

• Page 509-510: 17, 21
• For 17a show the full process--hypotheses, conditions, math, conclusion
• For 17b CALCULATE THE INTERVAL EVEN THOUGH YOU DID NOT HAVE EVIDENCE OF A DIFFERENCE
• you can just get the interval using your graphing calculator, copy your calculator screen, and then interpret this interval
• Consider the interval: how/why does your interval further support the idea that there "is not evidence of a decrease in voter support?"
• Answers not in the back:
• 17a.) Conditions: We know the first poll randomly selected voters; we'll assume the second poll used a random sample as well. Both 630 and 1010 are less than 10% of all voters in this state/town/district. (630)(.52)>10, (630)(1-0.52)>10, (1010)(0.52)>10, (1010)(1-.52)>10
• 17a.) Conclusion: The pvalue of 0.12 is above any common alpha level so we fail to reject the Ho. There is not evidence to suggest a decrease in voter support for his candidacy.
• 17b.) 99% Confidence Interval = (-0.0355, 0.09503)
• This interval does not suggest a significant difference (or a decrease) in voter support because the interval contains 0. If there was a decrease in support the entire interval would have been positive.
• 21a.) Conditions: We'll assume this is a representative sample of all women who never had mammograms and have had mammograms. 30,565 is likely less than <10 10="" 30131="" all="" had="" have="" is="" mammograms="" never="" of="" who="" women="">10% of all women who have never had mammograms and 30131 is likely less than 10% of all women who have had screening.  (30565)(.006) > 10,  (30565)(1-.006) > 10, (30131)(0.006) > 10, (30131)(1-.006) > 10

Another Stamp Opportunity (Stamps due next Thurs):

• You should have two leftover slides from our last chapter...one is about testing for ESP at Duke University and the other is about therapeutic touch.
• Stamp: Write the hypotheses for each scenario. Write the hypotheses on the slide (front or back).
• If you lost the slides you cannot earn these stamps. 

Today's Class Recap:

• Stamp = 2 prop z interval MC
• HW Questions?
• Randomization condition for #7...
• Ch. 22 Notes:
• Conducting a 2 Proportion Z Test....
• Hypotheses (notes--some new stuff here)
• Conditions (one change happens as a result of our assumption in the null)
• Math (2 prop z test on calculator, copy screen)
• What is p-hat pooled? Why do we use it? Where do we use it?
• Conclusion (same old--use pvalue to reject or fail to reject, there is or is not evidence, copy context from question)
• Connecting 2 prop z intervals and tests...
• How can we use an interval to decide if there is a significant difference in two proportions?

Here's the plan leading up to our next test:

• We have one more chapter in this unit about inference for proportions....
• Ch. 22 = Inferences for Differences in Proportions (2 Prop Z Tests and 2 Prop Z Intervals)
• Tuesday/Wednesday (3/19, 3/20): Ch. 20 Notes = 2 prop z-test and 2 prop z-int examples, notes
• Thursday (3/21): Finish ch. 22 notes, one more test/interval example/context
• Friday (3/22): Ch. 22 classwork
• Monday (3/25): Start ch. 23 (T Tests and Intervals)
• Tuesday (3/26): TEST!
• Test will be...
• Matching vocab for the entire unit
• Some FR and MC (based more heavily on the 2 prop z stuff)
• Thurs (3/28) - Fri (3/29): The t distribution
• Monday (4/1): Another T Test/Interval Given Data
• Tuesday (4/2): Ch. 23 Practice
• Weds (4/3): Ch. 23 Quiz, Start Paired T Tests and Intervals (ch. 25)
• Thurs (4/4) - Fri (4/5): Paired T Tests and Intervals (ch. 25)

2's Day HW!

Tonight we need to get some practice with our 2 proportion z intervals (chapter 22)!

Please complete the following in your textbook:

Page 508: 3, 7

• Be detailed/thorough! Do these as you would on the AP exam--that's why I'm only give you two!
• For 3d, consider your order of subtraction and the sign (+/-) of your interval....
• For 7a, consider if there was (random) assignment of treatments--if so, experiment. If not, observational study. And that's what you reference to explain.
• For 7b don't forget the full process--conditions, math, interpret! You will not earn full hw credit if you don't check conditions, show the math, and interpret your interval here.
• For 7c again consider the order of subtraction and sign of your interval....
• Here are the conditions and interpretation for #7b--they aren't in the back of the book....
• Conditions:
• Men were randomly assigned to have surgery or not
• 347 < 10% of all men who have surgery for prostate cancer
• 348 < 10% of all men who don't have surgery for this
• 16>10, 331>10; 31>10, 317>10
• The men were randomly assigned to treatments, so survival rates are independent for each group.
• Interpret:
• We are 95% confident that the true difference in rates of death for the two groups of men falls between 0.006 and 0.080 (Surgery - No) based on this study.   

Today's Class Recap:

• Stamp = compare two pictures
• Ch. 22 Notes:
• Why is it called a 2 proportion z interval?
• When do we use an interval/test?
• How do we create a 2 proportion z interval? (Completed an example in our notes)
• Conditions
• Math
• Why is the interval negative?
• Interpret the interval
• What does this interval actually mean?

Here's the plan leading up to our next test:

• We have one more chapter in this unit about inference for proportions....
• Ch. 22 = Inferences for Differences in Proportions (2 Prop Z Tests and 2 Prop Z Intervals)
• Tuesday/Wednesday (3/19, 3/20): Ch. 20 Notes = 2 prop z-test and 2 prop z-int examples, notes
• Thursday (3/21): Finish ch. 22 notes, one more test/interval example/context
• Friday (3/22): Ch. 22 classwork
• Monday (3/25): Finish ch. 22 and start ch. 23?
• Tuesday (3/26): TEST!
• Test will be...
• Matching vocab for the entire unit
• Some FR and MC (based more heavily on the 2 prop z stuff)

MONDAY!

If you were out today you must make up your test as soon as possible (before the start of Friday)! You can make up the test...

• Any free period/lunch (tomorrow, Weds, or Thurs)
• After school Weds. or Thurs.
• You will need the full period (46 minutes) so plan accordingly.

Tonight's HW = Get Stamps!

• These questions can be found in the Barron's Review Guide you were given on Friday.
• You must show all work to earn any stamps; these are all review. (You can use the answers to work backwards and "figure it out" if you need to, but no stamps for just copying those answers)
• This is all review!
• Complete these on separate paper to turn in tomorrow:
• Please write the page number and question number for each, then show your work and circle the final answer

• Page 77: 1 (write a sentence/provide a bullet explaining why the answer is correct)
• Page 83: 19
• Page 124: 17
• Page 205: 3
• Page 207: 11
• Page 209: 15
• Page 212-213: 29-32
• Page 252: 5
• Page 253: 8

• Check all answers using the Barron's book! NONE of your answers should be incorrect--if you got something wrong use the given solution to work backwards and "figure it out!"

THIS IS THE PLAN LEADING ALL THE WAY TO THE AP EXAM!

• We have one more chapter in this unit about inference for proportions....
• Ch. 22 = Inferences for Differences in Proportions (2 Prop Z Tests and 2 Prop Z Intervals)
• Tuesday/Wednesday (3/19, 3/20): Ch. 20 Notes = 2 prop z-test and 2 prop z-int examples, notes
• Thursday (3/21): Finish ch. 22 notes, one more test/interval example/context
• Friday (3/22): Ch. 22 classwork
• Monday (3/25): Finish ch. 22 and start ch. 23?
• Tuesday (3/26): TEST!
• Test will be...
• Matching vocab for the entire unit
• Some FR and MC (based more heavily on the 2 prop z stuff)
• Wednesday (3/27): SAT Day
• Thursday (3/28) and Friday (3/29): T Tests and Intervals (ch. 23)
• Monday (4/1) to Friday (4/5): ch. 23 notes, practice, and quiz (t tests and t intervals)
• Week of 4/8: April Break
• Thurs 4/11 = AP Stat Review Session
• Week of 4/15: Ch. 24: 2 Sample T Tests ans 2 Sample T Intervals
• Weds. 4/17 = AP Meeting (in cafe)
• Week of 4/22: Ch. 25: Paired T-Tests and T Intervals
• Week of 4/29: Ch. 26: Chi-Squared Tests
• *AP Exams Start*
• Week of 5/6: Ch. 27: T Tests and T Intervals for Slope AND Review
• 5/13, 5/14, 5/15: Ch. 27: T Tests and T Intervals for Slope AND Review
• Thursday, 5/16: AP STAT EXAM!

HW = STUDY!

TEST ON EVERYTHING "ONE PROP Z INT" AND EVERYTHING "ONE PROP Z TEST" RELATED IS MONDAY! COME READY!

How can I study?

1. Use your Barron's review guide (passed out today in class)!
• CI = p. 289
• Hyp. Tests = p. 333
• First you can skim through the summary of each topic (starting on the page numbers above)
• This book lumps all types of intervals/tests together, so remember that we are only focusing on one proportion z intervals and one proportion z tests for now
• Then--practice problems!
• After the summary "notes" you can find a bunch of practice MC and FR to do in the book, followed by the answers to each question WITH explanations of those answers!
• This is an amazing/awesome resource--use it!
2. Finish the 6 MC questions we started in class! Here's the key....
1. E
2. D
3. E
4. A
5. A
6. D
3. Do the "AP Stat Test Review: Ch. 19, 20, 21" worksheet provided below, then check your answers (also below!)

Come ready! Monday's test WILL be a challenge--it'll (of course) include AP MC and some AP FR! 

If it helps, here's a list of what you need to know:

• What should I know about Hypothesis Testing
• What is the (4 step) process for any hypothesis test?
• What is a null hypothesis? Alternative hypothesis?
• What is the general logic of a hypothesis test--what are we trying to do?
• How can I write a null and alternative hypothesis for a given context?
• How do we check the conditions for a one proportion z test? 
• Which condition differs for a one prop z test v. a one prop z interval?
• Why do we check this condition differently for each process?
• What math work do we need to show for a one proportion z test?
• What are we trying to do mathematically?
• How do we know if a result is statistically significant?
• How can we calculate a p-value? 
• Watch the AP Stat Guy Video about the logic of hypothesis testing!
• Logic of Hypothesis Testing (Click me!)
• What is a p-value? 
• How can we calculate a p-value? 
• Check out the AP Stat Guy video about p-values!
• What is a p-value? (Big focus for Monday's class) (Click me!)
• How do we write conclusions for a hypothesis test based on a p-value and alpha level?
• Linkage: connect pvalue to alpha (higher or lower) 
• Decision: reject or fail to reject Ho
• Context: there is or is not evidence of the Ha
• What is a one tailed test? A two tailed test?
• How do pvalues of one and two tailed tests relate? 
• What is a Type I error? A Type II error?
• How can we define a Type I or Type II error in context, AND identify a consequence of each type of error?
• What are alpha and beta in relation to the types of error? 
• How do alpha and beta relate? 
• How can we decrease the probabilities of both types of error? 
• What is power? How is power calculated?
• How can we increase the power of a test?

• What do I need to know about confidence intervals?
• When/why do we create a confidence interval?
• What symbol is a sample proportion/%? A population proportion %? (statistic v. parameter)
• How do we identify/calculate x, n, and p-hat for a given context? 
• What wording in a question suggests I have to create a CI?
• How do I find the critical value for a given C. Level using...
• t-table 
• invnorm(
• What is the full CI process?
• How do I check conditions for a one prop z interval?
• How do I do the math with my calculator and show the supporting formula? 
• How do we interpret the meaning of a confidence interval?
• How can we use our confidence intervals to assess some claim? (like 13d from the weekend hw)
• What is ME? 
• How can we calculate margin of error given a sample size, C. Level, and p-hat? 
• How can we calculate sample size for a given ME?
• How does changing n affect the ME? the width of an interval?
• How does changing C. Level (and as a result, z*) affect the ME? the width of an interval?
• How can we find the value of p-hat (a point estimate) only given an interval? 
• How can we find the ME only given an interval? 
• What does it mean to say "__% confident?"--how can we interpret the meaning of a confidence level? 

• If you have any questions, things you're confused about, or just want a deeper understanding, take the time this weekend to use your resources!
• Watch AP Stat Guy videos 6-9 in Unit 5 for some help!
• Click Me! AP Stat Guy Videos!
• Use Khan Academy--this is aligned with AP Stat!
• Linked on the right WITH a course code
• Google stuff--anything about one proportion z intervals!
• Barron's book
• Read above!

Thursday HW = Ch. 21 Wrap Up!

Tonight please complete the "Ch. 21: Types of Error" worksheet provided in class (or below) to wrap up chapter 21.

• (Tomorrow (in class) you'll get an answer key to check these all out)
• Periods D, E, and H: skip #1b (or try to figure it out), we did not address this in class
• Period A you have a table we discussed that is used for #1b

Tomorrow in class we'll do some practice AP MC/FR in groups to finish up this unit and to prepare for the big day--TEST MONDAY!

Also, don't forget: Extra Credit HW + Stamps (Due Tomorrow)

• Complete AND SCORE the 2012 AP Free response on the back of Monday's HW paper....
• Score using a different color pen/pencil
• Here is the link to the scoring rubric:
•  2012 Scoring Rubric--Scroll to question #5 (click me!)
• You will only earn extra credit/stamps if this assignment is completed AND scored (with notes/comments). If you do not score it and take notes on your scoring, nothing will be awarded.
• Scoring:
• Score each section (a, b, c) as E, P, or I AND provide bullets explaining why you got that score...
• If you earned a P, for example, write down notes about what you would need to add to earn an "E"

Today's Class Recap:

• Warmup......
• HW questions (from textbook)
• Finished Ch. 21 Notes:
• How does an alpha level relate to the types of error?
• How do the types of error relate to how we (would) choose an alpha level for a study?
• What is beta?
• How does changing alpha level affect beta?
• How can we decrease the probabilities of both types of error (alpha and beta)?
• What is power? How can we interpret power and how is it calculated?
• How can we increase the power of a test?
• Looked at an example for our notes....

Here's the worksheet for tonight's homework in case you need it:

Wednesday HW!

Tonight please complete the following in your textbook--some practice with the types of error stuff, and some great review for Monday's test....

Weds. HW: Page 491-494: 3, 7ab, 11abcd, 13, 19bcd

• For #3, first you have to decide if the original decision was "reject" or "fail to reject the null"--if the researcher concluded that a new device IS significantly better, did he/she reject or fail to reject?
• Now, based on that, what does that tell us about our p-value? How did the p-value compare to the alpha level of 0.05 (bigger or smaller?)
• And finally, given what we know about the p-value (bigger or smaller than 0.05), do we think the same decision would be made at 1% significance level? 10%? Explain...
•  For 7b use your interval to decide if you would reject or fail to reject the null, then state whether there is or is not evidence that the fraction has fallen below the 5% mark, and explain why (based on your interval)
• This is like the middle school bullying example....
• For 11d consider a consequence of each type of error--it might be easier to just describe a consequence of a Type 1 with b, and then a consequence of a Type II with c (and then you can skip d)

Extra Credit HW + Stamps (Due Friday)

• Complete AND SCORE the 2012 AP Free response on the back of Monday's HW paper....
• Score using a different color pen/pencil
• Here is the link to the scoring rubric:
•  2012 Scoring Rubric--Scroll to question #5 (click me!)
• You will only earn extra credit/stamps if this assignment is completed AND scored (with notes/comments). If you do not score it and take notes on your scoring, nothing will be awarded.
• Scoring:
• Score each section (a, b, c) as E, P, or I AND provide bullets explaining why you got that score...
• If you earned a P, for example, write down notes about what you would need to add to earn an "E"

Today's Class Recap:

• Ch. 20 Vocab quiz
• HW questions (for periods D, E, H)/Discuss #3, 4
• Ch. 21 Notes:
• Defining Types of Error (Type I and Type II Errors) AND describing a consequence of each
• Completed the "top right slide" about MA seatbelts as an example of this

Vocab Quiz Tomorrow + HW!

1.) Please complete the "testing hypotheses with an interval" problem (context = a media report about bullying behavior) provided in class or below. 

• Check out the red bullets in the image below--this is what I'm looking for in your answer to #3
• If you're having trouble deciding whether to reject or fail to reject the null based on your interval, first try to interpret the interval...
• What does your interval mean? 
• "We are ____% confident that the true ______________ falls between ____ and _____ based on this sample of ____."
• Now, based on that interpretation, do you think you should reject or fail to reject the null? Why? 
• Think: what were you trying to prove in this context? 

2.) Tomorrow we will start class with our chapter 20 vocab quiz! Here's what to study...

• Hypothesis Test (what do we use it for?--to see if a parameter has increased, decreased, or changed)
• Confidence Interval (what do we use it for?--to estimate the value of a population parameter)
• Sampling Variability
• Null Hypothesis
• Alternative Hypothesis
• Conditions: know how to check each, why we check each (from last vocab quiz), and which condition differs for an interval/test
• Standardized Test Statistic
• PValue
• Standard Error v. Standard Deviation
• Statistic
• Parameter
• Alpha Level/Significance Level
• Know how to write conclusions:
• When do we reject Ho (based on alpha level, p-value)?
• When do we fail to reject Ho (based on alpha level, p-value)?
• When do we say "there IS evidence of the Ha?".
• When do we say "there IS NOT evidence of the Ha?"
• Chapter 19 Vocab (that might be on this quiz)....
• Critical Value
• Point Estimate
• Interpret Confidence LEVEL v. Interpret Confidence INTERVAL
• Why do we check each condition?
• Randomization: to assume our sample represents the population
• 10%: to assume independence within our sample
• This means we want to assume that one person's response or one data value is not related to another
• Success/Failure: to make sure our sample is large enough

Today's Class Recap:

• Stamp = Understanding P-Values (like yesterday's packet)
• HW Questions + Connect to PValues
• Ch. 21 Notes:
• Types of Error (Type I and Type II Errors)
• Described what types of errors we could make using a trial context and/or using a "medical testing" context
• Looked at what alpha levels correspond to what p-values, based on whether we conduct a one or two sided test

And don't forget, we have a unit test on everything confidence intervals and hypothesis tests (that we've learned so far) on Monday 3/18!