Lesson 3

Question 1

What is a survey?

Answer 1

• An investigation in which information is systematically collected but in which the experimental method is not used.
• May be conducted by face-to-face inquiry, self-completed questionnaires, telephone, postal service, or in some other way.
• The generalizability of results depends upon the extent to which the surveyed population (sample) is representative.

Question 2

What is the purpose of surveys?

Answer 2

• Assess prevalence of disease - cross-sectional survey e.g. what is the prevalence of diabetes in Milton Keynes?
• Measure risk and protective factors of the respondent
• Measure outcomes
• Ad-hoc data – locally commissioned surveys provide an opportunity to collect information of interest not routinely collected

Question 3

What are advantages of prevalence studies?

Answer 3

Cheap and quick
Useful for healthcare planning and investigating trends over time
Useful when routine data not available

Question 4

What are disadvantages of prevalence studies?

Answer 4

Not usually useful for conditions with a short duration
Not really useful for investigating causality
Sampling and data collection need care

Question 5

Define ‘population’

Answer 5

the group of people in whom we are interested in and wish to apply the results of the survey to

Question 6

Define ‘sample’

Answer 6

group of individuals taken from larger population

Question 7

Define ‘sampling frame’

Answer 7

list of everyone in the population from whom sample taken

Question 8

Define ‘random (probability) sampling’

Answer 8

• everyone in the sampling frame has an equal probability of being chosen
• important to achieve a representative sample

Question 9

Define ‘non-random (non-probability) sampling’

Answer 9

• easier and convenient
• Unlikely to be representative
– Beware self-selecting samples!

Question 10

Explain simple random sampling

Answer 10

Each population member given an identifier and numbers selected at random

Question 11

Explain stratified sampling

Answer 11

Divide population into strata (subgroups) and select sample from each using simple random sampling

Question 12

Explain cluster sampling

Answer 12

Use natural ‘clusters’ in the population e.g. schools. Simple random sample of ‘clusters’ (e.g. schools). Study all individuals within clusters

Question 13

Explain systemic sampling

Answer 13

Every Nth population member selected

Question 14

Define validity

Answer 14

how well a test measures what it is purported to measure

the capacity of a test to give a true result

Question 15

Define repeatability/reliability

Answer 15

the degree to which a measurement made on one occasion agrees with the same measurement on a subsequent occasion

Question 16

Define random error

Answer 16

arandomimprecisionorvariable performance that is due to chance alone

Question 17

Define bias

Answer 17

a systematic error in sampling or measurement

Question 18

Define selection bias

Answer 18

error due to systematic differences in the characteristics of the groups being studied due to differences in the way they were selected

Question 19

What are two types of selection bias?

Answer 19

Sampling bias - non-representative sampling

Non-response bias - respondents differ from non-responders

Question 20

Define information (measurement) bias

Answer 20

error due to systematic differences in the measurement or classification of individuals in the groups being studied

Question 21

What are two types of information bias?

Answer 21

Instrument Bias: systematic error due to inadequate design, calibration or maintenance of instruments

Inter-Observer Bias: systematic error between measurements of different interviewers e.g. due to training

Question 22

Define ‘precision’

Answer 22

• if lots of variation (random error), precision is poor, if little random
variation, measurement is precise
• E.g. variation occurs in measuring blood pressure

Question 23

Define ‘accuracy’

Answer 23

• How close the average result is to the true value, poor accuracy is
systematic error (bias)
• E.g. sphygmomanometer not calibrated and reading 10mmHg too high

Question 24

What is ‘response rate’?

Answer 24

% of selected sample that take part in survey/study