Data and Graphical Summaries

Question 1

Observational studies

Answer 1

The investigators have no control over the subjects or quantities of interest; they are just observers.
The investigators cannot use randomisation for allocation into groups.

Question 2

association

Answer 2

that one thing is linked to another
easy to establish

Question 3

Does association prove causation?

Answer 3

No it does not, it may suggest it.

Question 4

Precautions in observational studies

Answer 4

(I) It is very difficult to establish causation
(II) Observational studies with a confounding variable can lead to Simpson’s Paradox
(III) Historical control

Question 5

Control

Answer 5

a subject who did not get the treatment.

Question 6

Controlling for confounders

Answer 6

trying to reduce the influence of confounding variables.

Question 7

contemporaneous control

Answer 7

Controlled experiments need to be performed in the same time period

Question 8

causation

Answer 8

the relationship between cause and effect

Question 9

Historical control

Answer 9

Sometimes time is a confounding variable - contemporaneous control

Question 10

Simpson’s paradox

Answer 10

The association between a pair of variables (X,Y) reverses sign upon conditioning of a third variable Z, regardless of the value taken by Z.

Sometimes there is a clear trend in individual groups of data that reverses when the groups are pooled together.

Question 11

IDA - Initial data analysis

Answer 11

First general look at the data, without formally answering the research questions
Helps you see whether the data can answer research questions
Identify the data’s main qualities
Suggest the population from which a sample derives

Question 12

IDA process

Answer 12

Data background: checking quality and integrity of the data
Data structure: what info has been collected?
Data wrangling: scraping, cleaning, tidying, reshaping, splitting, combining
Data summaries: graphical and numerical

Question 13

Variables

Answer 13

measures or describes some attributes of the subjects
Data with p variables is said to have dimension p

Question 14

Graphical summaries

Answer 14

Best highlight features of this data
To some extent we use trial and error

Question 15

Big data

Answer 15

Refers to massive amounts of data being collected
High dimensional: more variables than subjects
Requires more complex visualisations, complicated machinery

Question 16

Data

Answer 16

information about the set of subject being studied