theme 2 - data, figures and graphs Flashcards

1
Q

what is a binomial distribution? what is it based on?

A

-a distribution of binary data where there is only 2 outcomes e.g. pass or fail
-based on a fixed number of trials characterised by the probability of success in a single trial and the number of trials

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

what is a poisson distribution?

A

(similar to binomial but) measures the number of events / distribution of binary data from an infinite sample so gives probability of getting r events in a population whereas binomial distribution is a fixed sample so measures probability of getting r events in a trial

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

how is normal distribution different to binomial and poisson?

A

its for continuous data instead of discrete e.g. measuring the weights if newborn babies

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

why does a normal distribution usually form a bell shaped curve?

A

because most biological events are the consequence of multiple variables

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

what can standard deviation be used to determine?

A

the between subject variability or the subject variability

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

how do you calculate standard deviation?

A

sd = square root of variance

to find the variance:
1. find the mean of the data
2.subtract the mean from each data point
3. square each deviation
4. add all the squared deviations together
5. divide the sum by the number of data points in the population

then the sd is just square root of that

(this is for population sd, if youre doing sample sd then at the end you have to divide the sum by the number of data points MINUS 1)

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

why use a table to present data?

A

-concise and effective way to present large amounts of data
-data with differing units can be displayed
-precise values can be displayed

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

why use a boxplot instead of histograms?

A

-although both show variables divided into groups, box plots can show the spread of continuous data and any outliers

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

what are figure legends, what do they do, what do they need to contain?

A

-self contained text associated with each figure, containing all the info necessary to understand the figure
-they tell the reader what any abbreviations/symbols/colours ect mean
-they need a short title as well as the figure number

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

what are figures?

A

they are collections of graphs, tables and images grouped together to serve a particular purpose

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

what is quantitative data and give an example

A

numerical data such as the va score of a patient or height/weight

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

what is qualitative data and give an example

A

data that has been split into categories and is typically more descriptive e.g. a patients feedback on how well they felt their optometry appointment went

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

what is continuous data?

A

data that can be measured so can take any value e.g. length

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

what is discrete data?

A

count data e.g. number of students in a class/ shoe size

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

what is ratio data, giving an example and what is its opposite?

A

differences between measurements where a true zero exists so there can be no negative value in ratio data e.g. height and weight

interval data

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

what is interval data, give an example? what is its opposite?

A

differences between measurements but there is no true, measured on a scale where each point is placed at equal distance from one another e.g. temperature
ratio data

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

what is ordinal data, give an example and its opposite

A

data in ordered categories e.g. school grades like A, B, C
nomial data

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

what is nomial data, give an example and its opposite

A

data in categories with no ordering or direction e.g. hair colour
ordinal data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

what is binary data?

A

qualitative (categorical data) that has only 2 groups

18
Q

how is qualitative data displayed?

A

-as a proportion in a pie chart/ bar chart
-in a table

19
Q

what’s the point of mean median and mode (average) of continuous data?

A

it allows us to highlight the location of the dataset

20
Q

what is the point of range, IQR and SD when summarising continuous data?

A

it acts as a measure of dispersion/ variability

21
Q

what graphs are used to display continuous data?

A

histograms, scatter graphs, box plots

22
Q

how do you convert continuous data into categorical data? Give an example

A

-by dividing it up and changing it so it becomes dichotomised
-e.g. taking eye pressure values and categorising it into hypertension and normotension

23
Q

what is another word for relative frequency? what does it mean?

A

-proportion
-its how often something happens proportional to the total number of trials and is an estimate of probability

24
Q

when can relative frequency = incidence?

A

if it is the number of new cases over a given period

25
Q

when can relative frequency be referred to as prevalence?

A

when there is a number of existing cases at a specific time or over a given period

26
Q

what variables does prevalence depend on?

A

the incidence as well as duration

27
Q

what is the formula for risk?

A

risk = number of exposed people with the disease / total number of the exposed sample

28
Q

what words could you use instead of risk to be more ethical towards certain data sets e.g. a down syndrome screening

A

sensitivity/ chance due to ‘risk’ having negative connotations

29
Q

what is the formula for relative risk, what is another word for relative risk?

A

relative risk = risk in exposed group / risk in nonexposed group

relative risk = risk ratio

30
Q

what does relative risk allow you to do?

A

compare the risk between different groups e.g. what’s the difference in risk of acquiring AMD between smokers and non-smokers?

31
Q

what is the definition of attributable risk?

A

the incidence in the exposed group that can be attributed to the exposure

32
Q

what is the formulae for attributable risk and attributable risk as a %?

A

-attributable risk = risk in exposed group - risk in non-exposed group

-attributable risk percent = attributable risk / incidence of disease in exposed group

33
Q

What is the formula for NNH (number needed to harm)

A

NNH = 1/attributable risk

34
Q

how do you calculate NNT (number needed to treat)

A

NNT = 1/ attributable risk reduction

35
Q

what is NNH?

A

The number of individuals that need to be treated so that one individual presents an adverse reaction accountable to the treatment

36
Q

what is NNT

A

the number of patients you need to treat to prevent one additional bad outcome

37
Q

what is the formula to work out odds?

A

odds = number of exposed people with disease / number of exposed people without the disease

38
Q

what does odds ratio do?

A

it quantifies the association between two events (calculated in a similar way to relative risk)

39
Q

what is the formula for odds ratio?

A

odds ratio = odds in exposed group / odds in non exposed group

40
Q

how do you know whether to use relative risk or odds ratio for a study?

A

-If its a cohort study, it asks if the outcome is occurring by looking ahead so use relative risk
-if its a case-control study, it asks the frequency of exposure by looking back and so use odds ratio only

41
Q

how is quality of life measured/ quantified?

A

using
-time trade off assessment
or
-standard gamble assessment

42
Q

why do we measure quality of life?

A

to understand how a disease affects someone’s life or how a treatment improves a patient’s quality of life

43
Q

how is time trade of assessment done?

A

where the patient is asked to imagine they have 10 years left to live and decide how many of those years they’d give up - remaining time is expressed as a ratio between 0 and 1 and this is what is used to quantify QoL

44
Q

how is standard gamble assessment calulated?

A

the patient is asked to imagine they have 10 years left to live and asked how much they would give up. This is taken as a percentage and then subtracted from 1 to give the quality of life