When To Use Each Test
When to use each test? What is the sig value? Whether or not it is significant?
Read output and tell you what that value means (significant or not)
One sample: given population value and compare your sample to that population
Know if its significant: analyze compare means, one sample t-test
Independent sample t-test: lab one against lab five
Two independent groups (compare their means)
Significance use twotail
Define the correct groups:
Paired sample: only care about one lab and you look at before or after (temporal component) something has changed over time
Compare what you were like at the beginning of the class and what they were like after the class
ANOVA: test whether or not there is a significant difference between 3 different groups
P value by using a post-doc analysis using tukey
Tukey tests the comparison of each group (you get a p-value for each comparison)
Anova only tells us if there is a difference somewhere in there
Appropriate test to run: TUKEY (test individual groups to see if there is a difference) post-doc
Pvalue = 0.07 (there is no sig value because it is above 0.05)
If its above 0.05 there is no statistical difference
Run a one way ANOVA,
5 questions (more accurately 7 questions) and 1 bonus
Fun sheet will be given from both lab tests
There may be something small from the last test
Independent values
Analyze, compare means, gives you a window with variables, move over appropriate variable (age), put in whatever value
For a before and after data test, move over and put in appropriate mean variable (could be either variable)
Put dependent variable into the top box (test variable)
Move grouping variable to the other box, and define them
Question: using the independent data, compare the mean value of all groups to the age of 50? (Move over age)
Paired: we have a historical value for 10 for how people were before they took this course, we wanna compare that against the before value (move over before)
Group Age Before
Read output and tell you what that value means (significant or not)
One sample: given population value and compare your sample to that population
Know if its significant: analyze compare means, one sample t-test
Independent sample t-test: lab one against lab five
Two independent groups (compare their means)
Significance use twotail
Define the correct groups:
Paired sample: only care about one lab and you look at before or after (temporal component) something has changed over time
Compare what you were like at the beginning of the class and what they were like after the class
ANOVA: test whether or not there is a significant difference between 3 different groups
P value by using a post-doc analysis using tukey
Tukey tests the comparison of each group (you get a p-value for each comparison)
Anova only tells us if there is a difference somewhere in there
Appropriate test to run: TUKEY (test individual groups to see if there is a difference) post-doc
Pvalue = 0.07 (there is no sig value because it is above 0.05)
If its above 0.05 there is no statistical difference
Run a one way ANOVA,
5 questions (more accurately 7 questions) and 1 bonus
Fun sheet will be given from both lab tests
There may be something small from the last test
Independent values
Analyze, compare means, gives you a window with variables, move over appropriate variable (age), put in whatever value
For a before and after data test, move over and put in appropriate mean variable (could be either variable)
Put dependent variable into the top box (test variable)
Move grouping variable to the other box, and define them
Question: using the independent data, compare the mean value of all groups to the age of 50? (Move over age)
Paired: we have a historical value for 10 for how people were before they took this course, we wanna compare that against the before value (move over before)
Group Age Before