Review- Chapter 8 Statistics

0.0(0)
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
Card Sorting

1/21

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

22 Terms

1
New cards

Statistical inference

  • drawing a conclusion about a population parameter based on a sample statistic

2
New cards

Confidence Interval

  • used to estimate the value of a parameter with a range of plausible values

  • estimate population proportion(p) with a z test

  • " I am % confident that the true proportional of ALL is within the interval__to__)

  • Width of CI=1/āˆšn

  • CI increase ā†’width increase ā†’z* increase

  • CI decrease ā†’width decrease ā†’ME and SE decrease

  • MORE narrow graph

    • decrease CI or Increase n

  • Higher CI

    • accept a wider interval OR increase n

*MUST compare to population, not only the sample

1 Proportion Z test

3
New cards

Confidence Level

  • ā€œif you took ____samples and constructed the resulting confidence interval, the true proportion of ALL___ will be in __%(as a number) of these intervalsā€

    increase

4
New cards

Significance test

  • Assess the plausibility of a particular claim about a parameter

  • TEST claim about the population(p) with a z-interval

CONDITIONS

  1. Random sample(SRS)

  2. n1 p1 >=10, n1(1- p1 )>=10

  3. The sample size is <= 10% of the population size

  • condition of independence & not population >10n =okay!

  • the increase in variation due to the dependence almost offsets the decrease due to a relatively large sample

STEPS

Title

  1. Conditions

  2. Hypothesis, label

  3. Test statisticā†’pvalueā†’diagram(if need)

  4. Conclusion in context, doubts of validity

    1. Compare p-value with Ī±, level of significance

    2. compare test stat to z*

  • ā€œSince p value _Ī±, there is/is not statistically significant evidence to reject the null hypothesis, that is there is/is not statistically significant evidenceā€¦ā€™

<ul><li><p>Assess the <strong>plausibility</strong> of a particular <strong>claim</strong> about a parameter</p></li><li><p>TEST claim about the population(p) with a z-interval</p></li></ul><p>CONDITIONS</p><ol><li><p>Random sample(SRS)</p></li><li><p>n<sub>1 </sub>p<sub>1</sub> &gt;=10, n<sub>1(1- </sub>p<sub>1</sub> )&gt;=10</p></li><li><p>The sample size is &lt;= 10% of the population size</p></li></ol><ul><li><p>condition of independence &amp; <span style="color: red">not </span>population &gt;10n =<mark data-color="green">okay!</mark></p></li><li><p>the<span style="color: green"> increase</span> in <strong>variation</strong> due to the dependence almost <mark data-color="red">offsets</mark> the <span style="color: red">decrease</span> due to a relatively <span style="color: green">large</span> sample </p></li></ul><p>STEPS</p><p>Title</p><ol><li><p>Conditions</p></li><li><p>Hypothesis, label</p></li><li><p>Test statisticā†’pvalueā†’diagram(if need)</p></li><li><p>Conclusion in <strong>context</strong>, doubts of validity</p><ol><li><p>Compare p-value with Ī±, level of significance</p></li><li><p>compare test stat to z*</p></li></ol></li></ol><ul><li><p>ā€œSince p value _Ī±, there is/is not statistically significant evidence to reject the null hypothesis, that is there is/is not statistically significant evidenceā€¦ā€™</p></li></ul>
5
New cards

Pooled estimate

pĢ‚ =total success in BOTH sample size/total sample size

6
New cards

means and standard error of sampling distribution formula

knowt flashcard image
7
New cards

P value

  • probability of obtaining a sample statistics as EXTREME or MORE EXTREME than one obtained by the null hypotheses( Ho ) is assumed to be true

  • small(<0.05,0.1)ā†’ sufficient evidence to reject Ho

8
New cards

z*

  • Critical z

  • Ways to increase z*

    1. farther pĢ‚ from p

    2. sample size decrease

    3. po farther from 0.5

9
New cards

Margin of error

  • exactly Ā½ the width of CI

  • determined byā€¦

  • small MOEā†’ large n

  1. how MUCH statistic typically varies from parameter ĻƒpĢ‚

  2. how confident do we want our answer to be z*

MATH

z*āˆš((pĢ‚(1-pĢ‚))/n

use 0.5 for pĢ‚ if not given

<ul><li><p>exactly Ā½ the width of CI</p></li><li><p>determined byā€¦</p></li><li><p><span style="color: red">small</span> MOEā†’ <span style="color: green">large</span> n</p></li></ul><ol><li><p>how MUCH statistic typically <strong>varies</strong> from parameter Ļƒ<sub>pĢ‚</sub></p></li><li><p>how <strong>confident</strong> do we want our answer to be z*</p></li></ol><p>MATH</p><p>z*āˆš((pĢ‚(1-pĢ‚))/n</p><p>use 0.5 for pĢ‚ if not given</p>
10
New cards

Hypothesis

Ho = null hypothesis, must use an = sign

Ha= alternate hypothesis, can use <, >, <=,>=, ā‰ 

  • defines the parameter of interest ā†’ refers to the population

11
New cards

pĢ‚

sample proportion

12
New cards

po

assumed proportion in the null hypotheses(Ho ), the given in text

13
New cards

2 tail

ā‰ 

2p(z____)

14
New cards

1 tail

  • either < or >

  • in context, could be testing for strength, effectivenessā€¦

15
New cards

Errors

are inversely related

16
New cards

Type I error (Ī±)

  • rejecting the null hypotheses(Ho ) when it is true

  • increase if Ī± increase

17
New cards

Type II error(Ī²)

  • failing to reject null hypotheses(Ho )when its false

  • MORE likely for small n

    *the only way to have power

18
New cards

Power

(1-Ī²)

  • probability that type II error does not occur ā†’ correctly rejecting null hypotheses(Ho )when its false

  • influenced by Ī±

Power Increase byā€¦

  1. n Increase

  2. Ī± Increase

  3. SE decrease

  4. True parameter further from null

19
New cards

Difference of 2 Proportions

CONDITIONS

  1. 2 samples are taken randomly or randomly assigned

  1. n1pĢ‚1 , n1(1-pĢ‚1), n2pĢ‚2, n2(1-pĢ‚2) >=5

  2. The sample size is <= 10% of the population size

20
New cards

Confidence Interval of Difference of 2 Proportions

Captures 0

  • yes ā†’ __% confident that there is no difference between p1 and p2

    ā†’means that 1 value is positive, 1 value is negative

  • no ā†’ __% confident that there is a difference between p1 and p2

    ā†’not plausible for there to be no difference

  • n can sometimes be split in half

  • ā€œthe true difference in proportion between___and___isā€

  • 2 groups having the SAME ā€œproportionā€ā†’ risk ratio =1, so for a positive/greater proportion than the other CI range is the decimal +1.

<p>Captures 0</p><ul><li><p><span style="color: green">yes </span>ā†’ __% confident that there is <span style="color: red">no</span> difference between p1 and p2</p><p>ā†’means that 1 value is <span style="color: green">positive</span>, 1 value is <span style="color: red">negative</span></p></li><li><p><span style="color: red">no </span>ā†’ __% confident that there is a difference between p1 and p2</p><p>ā†’<span style="color: red">not </span><strong>plausible</strong> for there to be <span style="color: red">no </span>difference</p></li><li><p>n can sometimes be<strong> split in half</strong></p></li><li><p>ā€œthe<span style="color: green"> true</span> difference in proportion between___and___isā€</p></li><li><p>2 groups having the SAME ā€œproportionā€ā†’ <mark data-color="yellow">risk ratio </mark>=1, so for a <span>positive/greater </span>proportion than the other CI range is the decimal +1.</p></li></ul>
21
New cards

Test statistic of Difference of 2 Proportions

STEPS
name 2 proportion z test

result ā†’ C up, C Low

  1. Conditions

  2. Hypotheses

  3. Test statistic ā†’ p value ā†’ diagram(if need)

  4. Conclusion in context, doubts of validity

    1. compare p value and Ī±

<p>STEPS<br>name<mark data-color="yellow"> 2 proportion z test</mark></p><p>result ā†’ C up, C Low </p><ol><li><p>Conditions </p></li><li><p>Hypotheses</p></li><li><p><strong>Test statistic</strong> ā†’ p value ā†’ diagram(if need) </p></li><li><p>Conclusion in context, doubts of validity </p><ol><li><p>compare p value and Ī±</p></li></ol></li></ol>
22
New cards