Locations + Games + Narratives = whaiwhai (or something else)
Statistical Tests How to tell if something (or somethings) is different from something else.
-
date post
19-Dec-2015 -
Category
Documents
-
view
223 -
download
2
Transcript of Statistical Tests How to tell if something (or somethings) is different from something else.
![Page 1: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/1.jpg)
Statistical Tests
• How to tell if something (or somethings) is different from something else
![Page 2: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/2.jpg)
Populations vs. Samples
• Remember that a population is all the possible members of a category that we could measure– Examples:
• the heights of every male or every female• the temperature on every day since the
beginning of time• Ever person who ever has, and ever will, take a
particular drug
![Page 3: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/3.jpg)
Populations vs. Samples
• So a population is kind of abstract - typically you couldn’t ever hope to measure the entire population– Notable exceptions include:
• Standardized tests (mean IQ is 100 with std. dev. of 15)
• Special populations such as rare diseases or isolated groups of people
![Page 4: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/4.jpg)
Populations vs. Samples
• A sample is some subset of a population
– Examples:• The heights of 10 students picked at random• The participants in a drug trial
![Page 5: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/5.jpg)
Populations vs. Samples
• The notation– Sample statistics are usually regular letters
like s and – Population statistics are usually greek
letters like:
- the population mean
- the population standard deviation
X
![Page 6: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/6.jpg)
Populations vs. Samples
• Test your intuition:– Under what circumstances does the mean
of a sample equal the mean of the population from which it was drawn?
– What about the standard deviation?– What if your sample was very small relative
to the population?
![Page 7: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/7.jpg)
Populations vs. Samples
• Test your intuition:– Most importantly: What if you took more
than one sample
![Page 8: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/8.jpg)
Central Limit Theorem
• There is a distribution of sample means
![Page 9: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/9.jpg)
Central Limit Theorem
• There is a distribution of sample means
100
The population of IQ scores:
![Page 10: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/10.jpg)
Central Limit Theorem
• There is a distribution of sample means
100
The population of IQ scores:€
x = 95
Your Sample
![Page 11: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/11.jpg)
Central Limit Theorem
• There is a distribution of sample means
100
The population of IQ scores:€
x = 103
Your Sample
![Page 12: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/12.jpg)
Central Limit Theorem
• There is a distribution of sample means
100
The population of IQ scores:€
x = 99
Your Sample
![Page 13: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/13.jpg)
Central Limit Theorem
• There is a distribution of sample means
• This is the sampling distribution of the mean
![Page 14: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/14.jpg)
Central Limit Theorem
• What is the mean of the sampling distribution of the mean?
– mean of the sampling distribution approaches the mean of the population with many resamplings
€
x = u
![Page 15: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/15.jpg)
Central Limit Theorem
• What is the standard deviation of the sampling distribution of the mean?
– The standard error of the mean
€
X
=σ
n
Notice it will always be less than the standard deviation of the population!
![Page 16: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/16.jpg)
Central Limit Theorem
• What is the shape of the sampling distribution of the mean?
– Central Limit Theorem: the sampling distribution of the mean is normal regardless of the shape of the underlying distribution !
– This means you can use the Z transform and use the Z table
![Page 17: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/17.jpg)
The Logic of Statistical Tests
![Page 18: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/18.jpg)
Statistical Tests
• Consider a simple example:– you are testing the hypothesis that eating
walnuts makes people smarter by feeding walnuts to a group of 30 subjects and then testing their IQ
![Page 19: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/19.jpg)
Statistical Tests
• Consider a simple example:– you are testing the hypothesis that eating
walnuts makes people smarter by feeding walnuts to a group of 30 subjects and then testing their IQ
– If you are right, then eating walnuts will make the average IQ of your subjects be higher than the average IQ of all people (the population) since, mostly, those other people don’t eat walnuts much
![Page 20: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/20.jpg)
Statistical Tests
• Consider a simple example:– Put another way:
• Is this sample (entirely) of walnut eaters different from the population of mostly non-walnut-eaters
![Page 21: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/21.jpg)
Types of Errors
• There are two “mistakes” you could make:
![Page 22: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/22.jpg)
Types of Errors
• There are two “mistakes” you could make:
– Type I error or False-Positive - you decide the walnut treatment works when it doesn’t really
– Type II error or False-Negative - you decide the walnuts don’t work when really they do
![Page 23: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/23.jpg)
Types of Successes
• There are two ways to succeed:
– Hit or True-Positive: You decide the walnuts do make people smarter and, in fact, they really do
– Correct-Rejection or True-Negative: You decide the walnuts don’t work and, in fact they really don’t
![Page 24: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/24.jpg)
Outcome Matrix
Works Doesn’tWork
“Works” True PositiveType I
“Doesn’t Work” Type II True-
Negative
Actual Situation
Your Conclusion
![Page 25: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/25.jpg)
Statistical Tests
• Consider a simple example:
– Your subjects turn out to have a mean IQ of 107.5 (1/2 S.D. from the mean of the population) after eating walnuts
![Page 26: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/26.jpg)
Statistical Tests
• What are two reasons why the mean IQ of your subjects might be greater than the mean of the population?
1. you happened to pick 30 very smart people (i.e. university students)
– WARNING: Type I error is possible!
![Page 27: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/27.jpg)
Statistical Tests
• What are two reasons why the mean IQ of your subjects might be greater than the mean of the population?
1. you happened to pick 30 very smart people (i.e. university students)
– WARNING: Type I error is possible!
2. the walnuts worked
![Page 28: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/28.jpg)
Statistical Tests
• Usually we are worried about making a type I error so we need to know:
– What fraction of all possible groups of 30 subjects would have a mean IQ of 105 or less?
![Page 29: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/29.jpg)
Statistical Tests
• Usually we are worried about making a type I error so we need to know:
– What fraction of all possible groups of 30 subjects would have a mean IQ of 105 or less?
• In other words, we are interested not in the distribution of IQ scores themselves, but rather in the distribution of mean IQ scores for groups of 30 subjects
![Page 30: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/30.jpg)
The Z Test
…as it is more formally known
![Page 31: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/31.jpg)
Example Z Test
• Using our example in which we are testing the hypothesis that walnuts make people smarter
• null hypothesis is that they don’t
X = 107.5 = 100 = 15
![Page 32: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/32.jpg)
Example Z Test• Using our example in which we are testing the hypothesis that walnuts make people smarter (null hypothesis
was that they don’t)
• We want to know how many standard errors from the mean (of the sampling distribution of means) is 107.5
X = 107.5 = 15
€
=100
![Page 33: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/33.jpg)
Example Z Test
X = 107.5 = 15
€
Zx =x − μ x
σ x
Here’s what we’ve got:
Here’s what we can compute:
€
X
=σ
n
n = 30
That’s what we’re after so that we can use the Ztable
€
=ux =100
![Page 34: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/34.jpg)
Example Z Test
X = 107.5 = 15
Here’s what we’ve got:
Here’s what we can compute:
€
X
=σ
n=
15
30= 2.739
n = 30
€
=ux =100
Which is much less than 15!
![Page 35: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/35.jpg)
Example Z Test
X = 107.5 = 15
Here’s what we’ve got:
Here’s what we can compute:
n = 30
€
X
= 2.739
€
Zx =x − μ x
σ x
=107.5 −100
2.739= 2.738
€
=ux =100
![Page 36: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/36.jpg)
Example Z Test
X = 107.5 = 15
Here’s what we’ve got:
n = 30
€
X
= 2.739
€
Zx = 2.738
€
=ux =100
Thus X = 107.5 isn’t half a standard deviation from the sampling distribution mean!
It’s actually more than two and a half standard deviations from the sampling distribution mean !
![Page 37: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/37.jpg)
Example Z Test
X = 107.5 = 15
Here’s what we’ve got:
n = 30
€
X
= 2.739
€
Zx = 2.738
€
=ux =100
Looking up 2.739 in the Z table reveals that only .0031 or .31% of the means in the sampling distribution of mean IQs (for groups of 30 people each) would have a mean equal to or greater than 107.5!
![Page 38: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/38.jpg)
Example Z Test
• What this means is that you have only a 0.31% chance of making a type I error if you conclude that walnuts made your subjects smarter !
![Page 39: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/39.jpg)
Example Z Test
• What this means is that you have only a 0.31% chance of making a type I error if you conclude that walnuts made your subjects smarter !
• Put another way, there is only a 0.31% chance that this sample of IQs is taken from the regular population…walnut eaters are different
![Page 40: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/40.jpg)
Alpha
• Is .31% small enough? What risk of making a Type I error is too great?
![Page 41: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/41.jpg)
Alpha
• Is .31% small enough? What risk of making a Type I error is too great?
• There is no absolute answer - it depends entirely on the circumstances
![Page 42: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/42.jpg)
Alpha
• Is .31% small enough? What risk of making a Type I error is too great?
• There is no absolute answer - it depends entirely on the circumstances
• 5% or probability (p) = .05 is generally accepted
![Page 43: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/43.jpg)
Alpha
• Is .31% small enough? What risk of making a Type I error is too great?
• There is no absolute answer - it depends entirely on the circumstances
• 5% or probability (p) = .05 is generally accepted
• This rate of making Type I errors (ie. number of Type I errors per 100 experiments) is called the Alpha Level
![Page 44: Statistical Tests How to tell if something (or somethings) is different from something else.](https://reader035.fdocuments.us/reader035/viewer/2022062421/56649d385503460f94a122b3/html5/thumbnails/44.jpg)
Statistical Significance
• So we conclude that walnuts have a statistically significant effect on IQ with a probability of a Type I error of less than 5%
– In a research article we might say “the effect of walnuts on IQ was significant (one-tailed Z test, p = .0031)”