Differences Across Two Groups Flashcards

1
Q

What are the two possibilities when we have two samples that are different?

A

a) The two samples are different because of sampling variability
b) We have two samples where different is beyond sampling variability, so true difference

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

What are the pooled degrees for standard error, assuming equal standard deviation?

A

n1 + n2 - 2

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

Compare the degrees of freedom for assuming equal and unequal standard deviations

A

Always smaller degrees of freedom under unequal standard deviation, perceived as more conservative

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

What have we assumed about data when checking differences between two groups?

A
  • Independence of the data WITHIN and BETWEEN the groups

- The data are normally distributed

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

Why are hypothesis tests distinct from confidence intervals?

A

C.Interval contains a range of values for the parameter that are consistent with the data given a success rate
A hypothesis test tests one value is the true value

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

What are the three types of hypothesis?

A

Research Hypothesis
Null Hypothesis
Alternative Hypothesis

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

Describe a research hypothesis

A

Hypothesis that research is designed to investigate

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

Describe a null hypothesis

A

Hypothesis that we test and is typically a sceptical reaction to the research hypothesis

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

Describe an alternative hypothesis

A

Specifies a departure from H0, typically corresponsd to the research hypothesis

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

If H0 is true, should the test statistic be small or large?

A

Small

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

What do we do to decide if a test statistic is typically under the null hypothesis?

A

Compare our observed value with a reference distribution

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

Describe the p value

A

Probability of observing a test statistic at least as extreme as the one observed, given the null hypothesis is true

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

What does the p value measure?

A

The strength of evidence against the null hypothesis

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

What does a small p value indicate?

A

The test statistic is very unlikely to occur when the null hypothesis is true

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

What can we do when the p value is very small?

A

Reject H0 in favour of H1

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

Give examples of common threshold values for rejecting H0

A

0.1, 0.05 and 0.01

17
Q

What are the degrees of freedom for a one sample t test?

A

n - 1

18
Q

What is the non-parametric alternative to the t test?

A

Mann Whitney Test

19
Q

Describe the Mann Whitney Test

A
  • non parametric alternative to the t test
  • converts data to ranks
  • null : no difference between the ranks of the groups
  • alternative : difference between the ranks of each group
20
Q

What does a significant difference indicate?

A

Differences in group means are not likely due to sampling error