Session 4 - Lecture 1 - Intro to Cross-Sectional Survey

Question 1

2- What are the types of studies you can do?

Answer 1

• Ecological
• Cross-sectional
  (descriptive - studies distribution)
• Case-control
• Cohort
  (analytical - studies determinants)

^ these 4 are ‘watching’ studies

• Control trials
  (intervention)
  ^ where you ‘do’ something to someone and see results

Question 2

4 - What is population science?

Answer 2

The study of populations, i.e. >1 person. It is a networked idea.

Question 3

5 - What is the definition of epidemiology?

How can this be carried out?

Answer 3

Definition of Epidemiology
“Epidemiology is the study of the distribution and determinants of health-related states or events (including disease), and the application of this study to the control of diseases and other health problems.”

• Various methods can be used to carry out epidemiological investigations:
– surveillance and descriptive studies can be used to study distribution
– analytical studies are used to study determinants

Source: World Health Organisation. Health Topics: Epidemiology.
www.who.int/topics/epidemiology/en (20 March 2017)

{The study of the distrib & determinants of things related to health and applying that study to controlling diseases and other things relating to health.

Distrib – work out where it’s scattered
Determ - what leads to things, why do certain diseases happen.

Predominantly, simple terms – we do
- descriptive studies (ecological/cross-section) or
- analytical studies (case-control/cohort)
- 3rd ones are intervention studies, based on those studies, are control trials – based on doing things to people.
But the other ones are based on how we watch things – how we look & watch is how it actually matters.}

Question 4

6 - Why do we take a sample in epidemiological studies?

Answer 4

We often want to make inferences about the entire population… so we take a sample!
Generalise [picture of sample to population]

{We don’t study whole populations – even with big data – we only have a sample. The purpose is efficiency – why take a whole pop when you can take a sample to generalise to the whole population.}

Question 5

7 - What is the problem in taking a sample and not studying a whole population? Why does this happen?

Answer 5

Chance (Random error)
Sampling variation

• Frequency
• Groups [charts]

{But when you take a sample, and you do another one, and you do another one – it varies. It inevitably will vary}

Question 6

8 - What general factors do we need to consider in taking a sample?

Answer 6

Sample
We want our sample to be:
• Representative
by being
• Unbiased⇐ ‘Study Design’
and
• Precise ⇐ ‘Statistics’

{1. representative of the population that we were looking

• unbiased: if the population’s over here – you don’t want your sampling telling you it’s over there.
• But also don’t want too much uncertainty – want it as precise as you can

These 2 terms are completely different – more accurate result would give you 0 marks bc that could mean anything. But you need to distinguish the understanding between bias and precision.}

Question 7

9 - Explain what precision is and what an increase in precision does.

Answer 7

Chance (Random error)
Increase in precision/reduce uncertainty as sample size increases

• Frequency
• Sample_50
• Sample_250
• Sample_500
  Least Precise/Most uncertain –> Most Precise/Least uncertain

{Precision: as you get a bigger sample, you get more precise.}

Question 8

10

Answer 8

11

Question 9

11

Answer 9

12

Question 10

12

Answer 10

13

Question 11

13

Answer 11

14

Question 12

14

Answer 12

15

Question 13

15

Answer 13

16

Question 14

16

Answer 14

17

Question 15

17

Answer 15

18

Question 16

18/19

Answer 16

19

Question 17

19

Answer 17

19

Question 18

20

Answer 18

20

Question 19

21

Answer 19

21

Question 20

22

Answer 20

22

Question 21

23

Answer 21

23

Question 22

24

Answer 22

24

Question 23

25

Answer 23

25

Question 24

26

Answer 24

26

Question 25

27

Answer 25

27

Question 26

28

Answer 26

28

Question 27

29

Answer 27

29

Question 28

30

Answer 28

30

Question 29

31

Answer 29

31

Question 30

32

Answer 30

32

Question 31

33

Answer 31

33

Question 32

34

Answer 32

34

Question 33

35

Answer 33

35

Question 34

36

Answer 34

36

Question 35

37

Answer 35

37

Question 36

38

Answer 36

38

Question 37

39

Answer 37

39

Question 38

40

Answer 38

40

Question 39

41

Answer 39

41

Question 40

42

Answer 40

42

Question 41

43

Answer 41

43

Question 42

44

Answer 42

44

Question 43

45

Answer 43

45

Question 44

46

Answer 44

46

Question 45

47

Answer 45

47

Question 46

2- What are the types of studies you can do?

Answer 46

• Cross-sectional
• Case-control
• Cohort
• Ecological?