Statistics Flashcards

1
Q

Sensitivity calculation

A

True positives/number of people with disease

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

Specificity calculation

A

True negatives/number of people without the disease

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

Positive predictive value calculation

A

True positives/all positives

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

Negative predictive value calculation

A

True negatives/all negatives

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

Positive likelihood ratio calculation

A

Sensitivity/(1-specificity)

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

Negative likelihood ratio calculation

A

(1-sensitivity)/specificity

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

Variance calculation

A

(Value - mean) squared for each value and sum them
Divide by n-1

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

SD calculation

A

Square root of variance

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

What percentage of data is within 1 SD of mean?

A

68%

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

What percentage of data is within 2 SDs of mean?

A

95%

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

What percentage of data is within 3 SDs of mean?

A

99.7%

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

What is the coefficient of variation?

A

SD as a percentage of the mean

SD / mean x 100

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

SEM calculation

A

SD / square root of n
or
square root of SD squared / n

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

How do you normalise skewed data?

A

Take the log, square root, square, or reciprocal (1/x)

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

95% CI calculation

A

+/- SEM x 1.96

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

Calculation of SD in binomial sample

A

p = proportion e.g. 40%
p x (1-p)

17
Q

Calculation of SEM in binomial sample

A

square root of: p x (1-p) / n

18
Q

What is Poisson distribution?

A

Number of occurrences of an events per unit of time (distribution of discrete data)
Variance = mean

19
Q

SD calculation for Poisson distribution?

A

Squre root of mean
(mean = variance)

20
Q

Types of data

A

Nominal - naming scale e.g. eye colour
Ordinal - ordered values e.g. first, second, third
Discrete - integers
Continues - can be non-whole numbers
Interval - distance between values is the same and can be +ve or -ve e.g. 1, 2 ,3 or 5, 10, 15
Ratio - interval scale with fixed 0 e.g. height, weight

21
Q

Statistical test for correlation between two variables

A

Normal - Pearson’s coefficient
Non-normal - Spearman’s ranked coefficient

23
Q

Z score calculation?

A

(Value - mean) / SD

24
Q

Maternal mortality ratio

A

Maternal deaths per 100,000 live births in a year
(8.2 in UK in 2008)

25
Q

Maternal mortality rate

A

Maternal deaths per 1000 women of reproductive age

26
Q

Perinatal mortality rate

A

Number of stillbirths and deaths in first week of life per 1000 live births
(8 in 1000 in UK)

27
Q

Perinatal mortality rate

A

Number of stillbirths from 22/40 and deaths in first week of life per 1000 live births
(8 in 1000 in UK)