RESS - Revision Lecture

Question 1

What is epidemiology?

Answer 1

Epidemiology is the “study of the distribution & determinants of health related states or events in specified populations”

Question 2

Give examples of valid keyword search techniques

Answer 2

Truncation
Adjacency searching
Wildcards
Wildkeys

Question 3

What are truncations?

Answer 3

Truncation = uses the asterisk – type in the shared / common part of the word followed by the truncation symbol. E.g. anorexi* = anorexia, anorexic

Question 4

What is Adjacency Searching?

Answer 4

Adjacency searching: adj(n-1)
E.g. eating adj2 disorder = eating disorder, eating related disorder
ADJ1 searches for both terms next to each other in either order.
 ADJ2 searches for both terms and up to one word in between them
 ADJ3 searches for both terms and up to two words in between them
 ADJn searches for both terms and up to (n-1) words in between them

Question 5

What are wildcards?

Answer 5

Use a # or a ? in a word
 # is used to search for a variable single character, e.g. organi#ation will find organisation and
organization
 ? is used to search for a single character, or no character at all. This is very useful for some British and American spellings, e.g. orthop?edics will find both orthopaedics and orthopedics

Question 6

What is the boolean operation?

Answer 6

The us of AND OR and NOT in a search

Question 7

Examples of boolean operation?

Answer 7

AND will find anything with both words you are looking for
OR will find anorexia on its own, and CBT on its own but not necessarily both (i.e. looks for synonyms)
NOT will find anorexia where CBT isn’t mentioned

Question 8

How would you summarise the average of normally and non-normally distributed variable?

Answer 8

Normally: Mean
Non normally: Median
The mean is affected by abnormally high or low values, the median is not.

Question 9

What do SD and IQR measure?

Answer 9

Standard deviation and inter-quartile range are measures of spread. Use sd with mean, and IQR with median

Question 10

What are useful plots for distribution?

Answer 10

Histograms and bar charts;
NOT SCATTERPLOT (how variables are related) AND BOX PLOT (Median, IQr and outliers)

Question 11

What is person time?

Answer 11

A way of determining how many people are at risk. Use this as the incidence denominator

Question 12

What does the p value measure?

Answer 12

Probability of an event
(probability that the null hypothesis is true)
Smaller p value less likely

Question 13

P value of 0.05

Answer 13

So p=0.05 means a 1 in 20 (5%) chance of an event happening
Combined with a hypothesis , it’s a 5% chance of seeing the results you saw if the null hypothesis were true.
Small = reject null hypothesis

Question 14

What does the p value provide?

Answer 14

Evidence not proof and whether a statistical test was accurate

Question 15

What data type do you get if you use the Likert scale?

Answer 15

Ordinal as its a scale