gea1000 chp1

Question 1

PPDAC full form

Answer 1

Problem, Plan, Data, Analysis, Conclusion

Question 2

Population

Answer 2

The entire group of individuals or objects that we wish to know about

Question 3

Research question

Answer 3

Seeks to investigate some characteristic of a population

Question 4

Population of interest

Answer 4

A group in which we have interest in drawing conclusions in a study

Question 5

Population parameter

Answer 5

Numerical fact about a population

Question 6

Census

Answer 6

An attempt to reach out to the entire population of interest

Question 7

Drawbacks of a census

Answer 7

1. High cost of conducting
2. Takes a long time to complete - some studies are time sensitive
3. One may not be able to achieve 100% response rate

Question 8

Sample

Answer 8

Proportion of the population selected in the study

Question 9

Sampling frame

Answer 9

List from which the sample was obtained

Question 10

Conditions for generalisability

Answer 10

1. The sampling frame must be equal to or greater than the population of interest
2. There should be no bias when we obtain the sample (selection bias, non-response bias)

Question 11

Selection bias

Answer 11

• Associated with the researcher’s biased selection of units into the sample
• This can be caused by imperfect sampling frame, which excludes units from being selected
• Can also be caused by non-probability sampling

Question 12

Non-response bias

Answer 12

• Associated with participants’ non-disclosure or non-participation in the research study
• This results in the exclusion of info from this group
• E.g. inconvenience or unwillingness to disclose sensitive info

can occur regardless of whether the sampling method is probabilistic or non-probabilistic

Question 13

Probability sampling

Answer 13

A sampling scheme such that the selection process is done via a known randomised mechanism

Every unit in the sampling frame has a known non-zero probability of being selected but the probability of being selected doesn’t have to be the same for all units

Question 14

Simple random sampling

Answer 14

1. Units are randomly selected from the sampling frame
2. Every unit of the sampling frame has equal chance to be selected
3. Sampling without replacement

Question 15

Systematic sampling

Answer 15

A method of selecting units from a list by applying a selection interval k and a random starting point from the first interval

e.g. in an interval of k, the rth number is chosen where
1 <= r <= k
r, r+k, r+2k .. r + (n-1)k

Question 16

Pitfalls of systematic sampling

Answer 16

Potentially under-representing the population

Question 17

Stratified sampling

Answer 17

• Sampling frame is divided into groups called strata
• Each stratum is similar in that they share similar characteristics but the size of each stratum is not necessarily the same
• We apply SRS to each stratum to generate the overall sample

Question 18

Pitfalls of stratified sampling

Answer 18

Require sampling frame and criteria for classification of the population into stratum

Question 19

Cluster sampling

Answer 19

• Sampling frame is divided into clusters
• A fixed number of clusters are then selected using SRS
• All the units from the selected clusters are then included in the overall sample

Question 20

Advantages of cluster sampling

Answer 20

Less time-consuming and less costly
Clusters as usually naturally defined so it is easy to classify a unit under a cluster

Question 21

Disadvantage of cluster sampling

Answer 21

• Depending on which clusters are selected, we may see high variability in the overall sample if there are dissimilar clusters with distinct characteristics
• If the number of clusters sampled is small, there is also a risk that the clusters selected will not be representative of the population

Question 22

Pitfalls of simple random sampling

Answer 22

Time-consuming; accessibility of information and sampling frame

Question 23

Non-probability sampling

Answer 23

A non-probability sampling method is when the selection of units is not done by randomisation
There is no element of chance in determining which units are selected - down to human discretion

Question 24

Convenience sampling

Answer 24

Non-probability sampling method where a researcher chooses subjects based on the most easy availability
* introduces selection bias
* introduces non-response bias

Question 25

Volunteer sampling/Self-selected sampling

Answer 25

Subjects volunteer themselves into a sample * Sample contains subjects who have a stronger opinion on the research question than the rest of the population * Such a sample is unlikely to be representative of the population of interest

Question 26

Independent variables

Answer 26

Those that may be subject to manipulation, either deliberately or spontaneously, in a study

Question 27

Dependent variables

Answer 27

Those that are hypothesised to change depending on how the independent variable is manipulated in the study

Question 28

Categorical variables

Answer 28

Variables that take on categories or label values Ordinal: There is some natural ordering and numbers can be used to represent this ordering Nominal: There is no intrinsic ordering

Question 29

Numerical variables

Answer 29

Variables that take on numerical values and we can meaningfully perform arithmetic operations on them Discrete numerical variable: there are gaps in the set of possible numbers taken on by the variable. Continuous numerical variable: can take on all possible numerical values in a given range or interval.

Question 30

If we add a constant value c to all data points how does the mean of that dataset change?

Answer 30

mean + c

Question 31

If we multiply a constant value c to all data points how does the mean of that dataset change?

Answer 31

c * mean

Question 32

Variance formula Standard deviation formula

Answer 32

var = ((x₁ - mean)² + ... + (x_n - mean)²)/(n-1) standard deviation = sqrt(var)

Question 33

when is standard deviation 0

Answer 33

when all data points are identical in this case variance is 0 so standard deviation also is

Question 34

When a constant c is added to all data points how does standard deviation change

Answer 34

does not change

Question 35

When a constant c is multiplied to all data points how does standard deviation change

Answer 35

Standard deviation is multiplied by |c|

Question 36

Coefficient of variation

Answer 36

Used to quantify the degree of spread relative to the mean Formula = standard deviation/mean

Question 37

How does the median change when a constant c is added to every data point

Answer 37

median + c

Question 38

How does the median change if all data points are multiplied by c

Answer 38

median * c

Question 39

How does the IQR change if we add c to all points

Answer 39

No change

Question 40

How does the IQR change if we multiply each point by c

Answer 40

IQR is multiplied by |c|

Question 41

For numerical variables when is median and IQR preferred over mean and standard deviation?

Answer 41

The median and IQR is preferred if the distribution of the data is not symmetrical or when there are outliers

Question 42

Primary goal of an experimental study

Answer 42

To provide evidence for a cause-and-effect relationship between two variables

Question 43

What is a placebo

Answer 43

Something given to the control group that in actual fact has no effect on the subjects on the group

Question 44

What is an observational study

Answer 44

Measures the variables of interest without any direct/deliberate manipulation of the variables by researchers