Alpha {number}

number has to be below to accept null hypothesis?

if number above is to reject null hypothesis?

number compared with P - Value of Data Analysis

preposition to be tested;

the hypothesis under investigation

no theory

no statistical difference

"there's no statistical difference between mean set 1 and mean set 2."

do not use "increase" or "decrease" in terminology

analysis difference in multiple statistics.

key phrase: statistic - significant - difference

t - Test null - hypothesis analysis of 2 data to figure out if need improvement or no improvement.

Ex: ANOVA Tests bike racks on campus to see if each racks show significant distance from rack to destination. If 1 rack is too far from destination, then there's needs to have readjustment of rack to make it closer to destination.

distribution of occuring with 2 modes

means of # started high, but fall in middle and then high again, not good means = no # nothing in middle

Step 1

of testing

Step 2

of testing

Hypothesis

Step 3

of testing

Select Test Method

----------

Ex: Paired/Unpaired t - Test

Step 4

of testing

Step 5

of testing

Step 6

of testing

Step 7

of testing

P - Value of t - Test: 0.152

{Terminology Example}

there are equal means from any data sets

Ex: Column 1 = Column 2

P - Value of t - Test: 0.261

{Terminology Example}

We reject the null - hypothesis;

since there are no statistical signficant differences of the means between 2 data sets.

P - Value of t - Test: 0.038

{Terminology Example}

There are still statistical differences.

same place

before and after of the 2 sets

different places

2 different groups

EX: AZ & CA

non - directional hypothesis of any differences

between A & B

"t - critical"

one - tail & two - tail

Fail to reject the null - hypothesis

means?...

Correlation

{Term Definition}

Regression

{Term Def 1}

prediction & more details of data (extent X - axis)

testing for relationship NOT differences

----

X - input = independent

Y - input = Applicant (dependent)

Multiple R

{Regression}

R Squared (Multiple R)

{Regression}

co - efficient determination

percentage explain by model

"variation is R Squared"

negative value

Adjusted R Squared

{Regression}

the correlation of R squared and Adjusted R Squared

closer to the line the better

Significant F

(Significant Factor)

{Regression}

P - Value of Significant F = 3.68936 E - 06

{Terminology Example}

(higher than 0.05)

There is NOT statistical significant relationship between income and median house price.

We reject the null - hypothesis.

Using "Regression" Data Analysis

{what to look for}

1. P - Value NEEDS to be BELOW 0.05 of Significant F
2. If above = no SSR = not valid = reject null
3. If below = there IS SSR between dependent & independetn variables
4. Terminology Example:
• For every point goes up, the student quality goes up by 1(X-input/indepedent) as well

Note: 4.68E-15 of P - Value is very low (bad)

Difference

{Term Definition}

Relationship

{Term Definition}

t - Test

{Term Def}

Regression

{Term Def}