Test 1 Flashcards

Study for Test 1

1
Q

What are the five steps of hypothesis testing?

A

(1) State hypothesis (2) State alpha (3) Calculate statisitic (4) Find p-value (5) Draw conclusion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is single sample t-test?

A

t=(xbar-mu)/(s/sqrt(n))

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are the assumptions of the t-test?

A

That the data come from a normal distribution

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the two sample z-test (when variances are considered different)?

A

z=(ybar1-ybar2)/sqrt((sigma12/n1)+(sigma22/n2))

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Why are the variances added?

A

Because variance cannot be negative: var(ybar1-ybar2)=var(ybar1)+var(ybar2)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the pooled variance?

A

((n1-1)*s12 + (n2-1)*s22)/(n1+n2-2)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the t-test when variances are considered the same?

A

t=(ybar1-ybar2)/(sp*sqrt(1/n1+1/n2))

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What are the R commands for finding p-values?

A

Two sided: 2*pnorm() Lt: pnorm() Gt: 1-pnorm()

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the test for equal/unequal variances?

A

F=s12/s22 where F follows a F(n1-1,n2-1) distribution and H0:sigma12==sigma22 and HA: sigma12 ne sigma22

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What are the assumptions for ANOVA?

A

(1) Data comes from normal distribution (2) Two population variances are the same

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What type of design is expected in ANOVA?

A

CRD or Completely Randomized Design

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How to decompose the variation in ANOVA?

A

Total variability is measured by the total sum of squares: SST=Sum for all groups i through alpha, the sum for all individuals j through n, (yij - ybar..)**2

SST=SStreatments+SSerror

Sum bla i thru alpha, sum bla j thru n, (ybari - ybar.. + yij - ybari.)

The ybari - ybar.. is the SStreaments

The ybarij - ybari. is the SSerror

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How to find the degrees of freedom for an ANOVA?

A

df for SStreatments is a-1

df for SSerror is n-a

df for SST is n-1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Is there a two-tailed test for an ANOVA?

A

No, F-tests are only one sided so use :

1-pf(f, df of numerator, df of denominator)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the connection between t-test and ANOVA?

A

F is t**2 where t is a two-sided test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

How to check the normality assumption?

A

Use qqnorm and qqline in R to do a normal probability plot

17
Q

How to test equal variance assumption in R?

A

var.test(x,y)