statistical analysis & techniques
DESCRIPTION
Statistical Analysis & Techniques. Ali Alkhafaji & Brian Grey. Outline. Analysis How to understand your results Data Preparation Log, check, store and transform the data Exploring and Organizing Detect patterns in your data Conclusion Validity Degree to which results are valid - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/1.jpg)
Statistical Analysis &Techniques
Ali Alkhafaji & Brian Grey
![Page 2: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/2.jpg)
Outline- Analysis
- How to understand your results- Data Preparation
- Log, check, store and transform the data
- Exploring and Organizing- Detect patterns in your data
- Conclusion Validity- Degree to which results are valid
- Descriptive Statistics- Distribution, tendencies & dispersion
- Inferential Statistics- Statistical models
![Page 3: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/3.jpg)
Analysis- What is “Analysis”?
- What do you make of this data?
- What are the major steps of Analysis?
- Data Preparation, Descriptive Statistics & Inferential Statistics
- How do we relate the analysis section to the research problems?
![Page 4: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/4.jpg)
Data Preparation- Logging the data
- Multiple sources- Procedure in place
- Checking the data- Readable responses?- Important questions answered?- Complete responses?- Contextual information correct?
![Page 5: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/5.jpg)
- Store the data- DB or statistical program (SAS, Minitab)- Easily accessible- Codebook (what, where, how)- Double entry procedure
- Transform the data- Missing values- Item reversal- Categories
Data Preparation
![Page 6: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/6.jpg)
Exploring and Organizing- What does it mean to explore
and organize?- Pattern Detection
- Why is that important?- Helps find patterns an
automated procedure wouldn’t
![Page 7: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/7.jpg)
Pattern Detection Exercise
- We have these test scores as our results for 11 children:
- Ruth : 96, Robert: 60, Chuck: 68, Margaret: 88, Tom: 56, Mary: 92, Ralph: 64, Bill: 72, Alice: 80, Adam: 76, Kathy: 84.
![Page 8: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/8.jpg)
Pattern Detection Exercise
- Now let’s try it together:
![Page 9: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/9.jpg)
Observations- Symmetrical pattern:
B G B B G G G B B G B
- Now look at the scores: Adam 76 Mary 92 Alice 80 Ralph 64 Bill 72 Robert 60 Chuck 68 Ruth 96 Kathy 84 Tom 56 Margaret 88
![Page 10: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/10.jpg)
Observations- Now let’s break them down by
gender and keep the alphabetical ordering:
Boys Girls Adam 76 Alice 80
Bill 72 Kathy 84 Chuck 68 Margaret 88 Ralph 64 Mary 92 Robert 60 Ruth 96 Tom 56
![Page 11: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/11.jpg)
1 2 3 4 540
50
60
70
80
90
100
Girls Boys
Order in List
Test
Sco
reObservations
![Page 12: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/12.jpg)
Conclusion Validity- What is Validity?
- Degree or extent our study measures what it intends to measure
- What are the types of Validity?- Conclusion Validity- Construct Validity- Internal Validity- External Validity
![Page 13: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/13.jpg)
- What is a threat?- Factor leading to an incorrect conclusion
- What is the most common threat to Conclusion Validity?
- Not finding an existing relationship- Finding a non-existing relationship
- How to improve Conclusion Validity?
- More Data- More Reliability (e.g. more questions)- Better Implementation (e.g. better protocol)
Threats toConclusion Validity
![Page 14: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/14.jpg)
Questions?
![Page 15: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/15.jpg)
Statistical Tools- Statistics 101 in 40 minutes
- What is Statistics?- Descriptive vs. Inferential
Statistics- Key Descriptive Statistics- Key Inferential Statistics
![Page 16: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/16.jpg)
What is Statistics?- “The science of collecting,
organizing, and interpreting numerical facts [data]”
Moore & McCabe. (1999). Introduction to the Practice of Statistics, Third Edition.
- A tool for making sense of a large population of data which would be otherwise difficult to comprehend.
![Page 17: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/17.jpg)
4 Types of Data- Nominal
- Ordinal
- Interval
- Ratio
![Page 18: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/18.jpg)
4 Types of Data- Nominal
- Gender, Party affiliation- Ordinal
- Interval
- Ratio
![Page 19: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/19.jpg)
4 Types of Data- Nominal
- Gender, Party affiliation- Ordinal
- Any ranking of any kind- Interval
- Ratio
![Page 20: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/20.jpg)
4 Types of Data- Nominal
- Gender, Party affiliation- Ordinal
- Any ranking of any kind- Interval
- Temperature (excluding Kelvin)- Ratio
![Page 21: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/21.jpg)
4 Types of Data- Nominal
- Gender, Party affiliation- Ordinal
- Any ranking of any kind- Interval
- Temperature (excluding Kelvin)- Ratio
- Income, GPA, Kelvin Scale, Length
![Page 22: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/22.jpg)
4 Types of Data- Nominal
- Gender, Party affiliation- Ordinal
- Any ranking of any kind- Interval
- Temperature (excluding Kelvin)- Ratio
- Income, GPA, Kelvin Scale, Length
![Page 23: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/23.jpg)
Descriptive vs. Inferential- Descriptive Statistics
- Describe a body of data- Inferential Statistics
- Allows us to make inferences about populations of data
- Allows estimations of populations from samples
- Hypothesis testing
![Page 24: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/24.jpg)
Descriptive Statistics- How can we describe a population?
- Center- Spread- Shape- Correlation between variables
![Page 25: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/25.jpg)
Average: A Digression- Center point is often described as
the “average”- Mean?- Median?- Mode?- Which Mean?
![Page 26: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/26.jpg)
Average: A Digression- Mode
- Most frequently occurring- Median
- “Center” value of an ordered list
![Page 27: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/27.jpg)
Average: A Digression- Mean (Arithmetic Mean)
![Page 28: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/28.jpg)
Average: A Digression- Mean (Arithmetic Mean)
nxi
![Page 29: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/29.jpg)
Average: A Digression- Mean (Arithmetic Mean)
- Geometric Meannxi
![Page 30: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/30.jpg)
Average: A Digression- Mean (Arithmetic Mean)
- Geometric Mean
- Used for Growth Curves
nxi
nix
![Page 31: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/31.jpg)
Center
- Mode
- Median
- Arithmetic Mean
- Geometric Mean
56 56 64 68 72 76 80 84 88 92 96 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 32: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/32.jpg)
Center
- Mode56
- Median
- Arithmetic Mean
- Geometric Mean
56 56 64 68 72 76 80 84 88 92 96 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 33: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/33.jpg)
Center
- Mode56
- Median76
- Arithmetic Mean
- Geometric Mean
56 56 64 68 72 76 80 84 88 92 96 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 34: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/34.jpg)
Center
- Mode56
- Median76
- Arithmetic Mean≈75.64
- Geometric Mean
56 56 64 68 72 76 80 84 88 92 96 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 35: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/35.jpg)
Center
- Mode56
- Median76
- Arithmetic Mean≈75.64
- Geometric Mean≈74.45
56 56 64 68 72 76 80 84 88 92 96 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 36: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/36.jpg)
Spread- Range
- High Value – Low Value- IQR
- Quartile 3 – Quartile 1- Quartile is “Median of the
Median”
![Page 37: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/37.jpg)
Spread- Variance
![Page 38: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/38.jpg)
Spread- Variance
nx
V i
22 )(
![Page 39: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/39.jpg)
Spread- Variance
- Standard Deviation
nxi
2)(
nx
V i
22 )(
![Page 40: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/40.jpg)
Variance & St Dev
- Mean
- Variance
- Standard Deviation
10 10 10 10 10 10 10 10 10 10 10 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 41: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/41.jpg)
Variance & St Dev
- Mean10
- Variance
- Standard Deviation
10 10 10 10 10 10 10 10 10 10 10 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 42: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/42.jpg)
Variance & St Dev
- Mean10
- Variance0
- Standard Deviation0
10 10 10 10 10 10 10 10 10 10 10 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 43: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/43.jpg)
Variance & St Dev
- Mean10
- Variance
- Standard Deviation
8 12 8 12 8 12 8 12 8 12 10 .
1 2 3 4 5 6 7 8 9 10 11.
![Page 44: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/44.jpg)
Variance & St Dev
- Mean10
- Variance
- Standard Deviation
8 12 8 12 8 12 8 12 8 12 10 .
1 2 3 4 5 6 7 8 9 10 11.
64.31140
![Page 45: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/45.jpg)
Variance & St Dev
- Mean10
- Variance
- Standard Deviation
8 12 8 12 8 12 8 12 8 12 10 .
1 2 3 4 5 6 7 8 9 10 11.
64.31140
91.11140
![Page 46: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/46.jpg)
Shape
from wikipedia.org
![Page 47: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/47.jpg)
Shape
from wikipedia.org
from free-books-online.org
![Page 48: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/48.jpg)
Correlation
![Page 49: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/49.jpg)
Correlation- Measure of how two (or more)
variables vary together- Holds a value from -1 to 1
![Page 50: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/50.jpg)
Correlation- Measure of how two (or more)
variables vary together- Holds a value from -1 to 1- Pearson Correlation Coefficient
from wikipedia.org
![Page 51: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/51.jpg)
Inferential Statistics- Allows us to make inferences about
populations of data- Allows estimations of populations
from samples- Hypothesis testing- Uses descriptive statistics to do all
of this
![Page 52: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/52.jpg)
Central Theorem of Statistics
- Central Limit Theorem- By randomly measuring a subset
of a population, we can make estimates about the population
- Law of Large Numbers- The mean of the results obtained
from a large number of samples should be close to the expected value
![Page 53: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/53.jpg)
Hypothesis Testing- Used to determine if a sample is
likely due to randomness in a population
- Null Hypothesis (H0)
- Alternative Hypothesis (Ha)
![Page 54: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/54.jpg)
Hypothesis Testing- Used to determine if a sample is
likely due to randomness in a population
- Null Hypothesis (H0)- The sample can occur by chance
- Alternative Hypothesis (Ha)
![Page 55: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/55.jpg)
Hypothesis Testing- Used to determine if a sample is
likely due to randomness in a population
- Null Hypothesis (H0)- The sample can occur by chance
- Alternative Hypothesis (Ha)- The sample is not due to chance
in this population
![Page 56: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/56.jpg)
Statistical SymbolsPopulation Name Sample
μ Mean xσ Standard
Dev sP Probability pN Number n
![Page 57: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/57.jpg)
Z-Test (Significance Testing)
![Page 58: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/58.jpg)
Z-Test (Significance Testing)
- Determines how many standard deviations the sample is away from the mean in the normal distribution
n
xz
![Page 59: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/59.jpg)
An Example- The average weight for an
American man is 191 lbs and 95% of all men weigh between 171 and 211 lbs. Suppose you want to determine the effect of eating fast food 3 times a week or more on men’s weight.
- A sample of 25 men who eat fast food 3 times a week (or more) have an average weight of 195 lbs.
![Page 60: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/60.jpg)
An Example- H0:
![Page 61: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/61.jpg)
An Example- H0: μ = 191
![Page 62: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/62.jpg)
An Example- H0: μ = 191- Ha:
![Page 63: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/63.jpg)
An Example- H0: μ = 191- Ha: μ > 191
![Page 64: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/64.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ =
![Page 65: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/65.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10
![Page 66: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/66.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10- =x
![Page 67: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/67.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10- = 195x
![Page 68: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/68.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10- = 195- n =x
![Page 69: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/69.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10- = 195- n = 25x
![Page 70: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/70.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10- = 195- n = 25
n
xz
x
![Page 71: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/71.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10- = 195- n = 25
n
xz
x
2510
191195
![Page 72: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/72.jpg)
An Example- H0: μ = 191- Ha: μ > 191- σ = 10- = 195- n = 25
n
xz
x
2510
191195
5104
2
![Page 73: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/73.jpg)
An Example
from wikipedia.org
![Page 74: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/74.jpg)
α Values- p = .023- Is this significant?- Dependent on α value
- Specified before sampling- Typically, no looser than α = .05- α = .025, .01, .001 are common- Smaller α value lead to greater
statistical significance
![Page 75: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/75.jpg)
Back to the Example- H0: μ = 191- Ha: μ > 191- p = .023- We can reject H0 if α = .05 or α
= .025- We cannot reject H0 if α < .023
![Page 76: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/76.jpg)
Error Types- Type I Error/False Positive
- Erroneously rejecting H0 and accepting Ha
- Probability = α- Type II Error/False Negative
- Erroneously failing to reject H0- Reducing the odds of one type of
error increases the odds of the other
![Page 77: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/77.jpg)
The Problem with Z-Testing
n
xz
![Page 78: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/78.jpg)
The Problem with Z-Testing
- σ is the population standard deviation
n
xz
![Page 79: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/79.jpg)
The Problem with Z-Testing
- σ is the population standard deviation
- Standard Error is used instead
- Using SE instead of σ results in a Student’s t test
n
xz
nsSE
![Page 80: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/80.jpg)
Confidence Interval
![Page 81: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/81.jpg)
Confidence Interval- Allows an estimation of the
population mean
- z* is the z-value where a given proportion of the data are between z and -z
- Student’s T-Test has an equivalent confidence interval
nzx
*
![Page 82: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/82.jpg)
Confidence Interval- Using our earlier example, we are
95% certain that the average weight of someone who eats fast food three times a week or more is in the range:
nzxCI
*
![Page 83: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/83.jpg)
Confidence Interval- Using our earlier example, we are
95% certain that the average weight of someone who eats fast food three times a week or more is in the range:
25102195*
nzxCI
![Page 84: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/84.jpg)
Confidence Interval- Using our earlier example, we are
95% certain that the average weight of someone who eats fast food three times a week or more is in the range:
25102195*
nzxCI
5102195
![Page 85: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/85.jpg)
Confidence Interval- Using our earlier example, we are
95% certain that the average weight of someone who eats fast food three times a week or more is in the range:
25102195*
nzxCI
)199,191(5
102195
![Page 86: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/86.jpg)
Confidence Interval- Using our earlier example, we are
99% certain that the average weight of someone who eats fast food three times a week or more is in the range:
251057.2195*
nzxCI
)2.200,8.189(5
1057.2195
![Page 87: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/87.jpg)
Confidence Interval- Using our earlier example, we are
90% certain that the average weight of someone who eats fast food three times a week or more is in the range:
251064.1195*
nzxCI
)3.198,7.191(5
1064.1195
![Page 88: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/88.jpg)
Other Inferential Stats- ANOVA (Analysis of Variance)
- Examines 3+ means by comparing variances across and within groups.
- Allows detection of interaction of means.
- Regression- Examines how well 1+ variables
predict other variables.- Generates an equation to allow
prediction.
![Page 89: Statistical Analysis & Techniques](https://reader035.vdocuments.mx/reader035/viewer/2022062323/5681632e550346895dd3a971/html5/thumbnails/89.jpg)
Questions?