RCT (Analysing the data from each trial arm to see if the outcome was present or not) Flashcards

1
Q

What should quantitative data be?

A

It should either be primarily in a numerical form or able to be transformed into a numerical form

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

What can quantitative data sometimes be and how to resolve this?

A

Quantitative data can be very large numerical data and so needs to be reduced down used descriptive statistics.

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

What are descriptive statistics?

A

Are ways to summarise and display quantitative dataa

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

What are the 4 different levels of measurement (the quantitative date will in one of these formats)?

A

1) Nominal
2) Ordinal
3) Interval
4) Ratio

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

How can quantitative data be presented?

A

1) Tables
2) Charts
3) Measure of central tendency
4) Measure of dispersion

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

What are the 3 measures of central tendency?

A

Mode - most frequent value
Median - central value
Mean - Arithmetic average

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

What are the 3 measures of dispersion?

A

1) Range - the difference between the highest and lowest value
2) Inter-quartile range - the difference between the value at 25% and 75%
3) Standard deviation - subtracting the ean from each value, then finding out the mean of the remaining values
4) Variance - SD x SD

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

Which 2 levels of measurement can use the measures of dispersion to display the data?

A

ONLY interval or ratio

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

Why do we need to statistically analyse the data?

A

In order to draw inferences from them in which we can generalise to the rest of the target population

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

What are the 2 ways to statistically analyse the data?

A

1) Hypothesis testing (using the P value)

2) Estimation (using confidence intervals)

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

What are the 7 steps in hypothesis testing?

A

1) Obtain null hypothesis
2) Obtain alternative hypothesis
3) Carry out the significance test
4) Obtain the test statistics
5) Compare the test statistics with the hypothesised critical value.
6) Obtain P value
7) Decision - can the null hypothesis be rejected?

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

What is the P value?

A

It is the probability that the data from the study allows us to ACCEPT the null hypothesis. Therefore, stating there is NOT a link between the IV and DV.

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

How do we know whether to accept or reject the null hypothesis using the P value?

A

The closer the P value is to 0, the more likely we can reject the null hypothesis. The closer it is to 1, we must accept the null hypothesis.

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

What is the statistical significance?

A

5%. Once the test statistic has been compared to the hypothesised critical value and the value obtained is equal or less than 0.05 - then the null hypothesis can be rejected and the alternative hypothesis can be accepted.

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

What is an alternative hypothesis?

A

A hypothesis which states there is a difference between the IV and DV, but does not state the direction of this difference.

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

What is the ‘power’ of a study?

A

It is represented as a percentage. It is an indication of the power the trial has to be able to detect differences between the trial arms

17
Q

What are the 2 disadvantages of the ‘P value’ ?

A

Doesn’t tell you about:

1) Effect size
2) The clinical significance of the results.

18
Q

What is a Type 1 effect?

A

When you reject the null (there is a difference) when in fact the null should have been accepted (as there actually wasn’t a difference)

19
Q

What is a Type 2 effect?

A

When you accept the null (there is no difference) when in fact the null should have been rejected (as there is a difference)

20
Q

What is a confidence interval?

A

A estimated range of values is calculated from the data. The confidence interval will be able to determine how confident we are that the unknown population parameter is within that range.

21
Q

How do we know if there is a difference between the 2 trial arms using the confidence intervals?

A
95% = the unknown population parameter is estimated to be 95% of this estimated range and there is a difference between the trial arms
10% = the unknown population parameter is estimated to be seen 10% of the estimated range and so it is unlikely there is a difference between the trial arms.
22
Q

Why are confidence intervals better than ‘P values’?

A

They can be used to determine the results:

1) Effect size
2) Their clinical significance