Data Collection Flashcards
Define census
A way of sampling such that information from the whole target population is obtained
Define sample
A subset of the target population
Define target population
Collection of objects/individuals we want to learn something about
What are the three general ways to collect data?
- Polls and surveys
- Experiments
- Observational studies
Define a poll/sample
Process of collecting data from a sample or subset of the population
What are two potential problems of a census?
- Difficult to make sure everyone participates
2. Some people may struggle to complete the form/questionnaire etc
Why sample over doing a census?
- More practical
- Faster
- Cheaper than census
Define a variable
Some characteristic of each individual in the population
Define a parameter
Numerical summary of a variable for a population
Define a statistic
Numerical summary of a variable for the sample
What is precision?
Implies that the value of the statistic is similar in all samples
What is bias?
Implies that the sample statistic tends to differ from the population parameter in a consistent way
What is the goal when taking the sample?
To have a sample that reflects the sort of variation that the whole population exhibits
What are the advantages of simple random sampling?
- Easier to draw, only one random number required
- Distributes the sample more evenly through the population
What is sampling error?
Difference between sample statistic and true but unknown parameter
Describe the steps for systematic sampling with random start
- Decide how many individuals you want in your sample, n
- Specify/estimate the size of your population, N
- Calculate fixed periodic interval, k = N/n
- Randomly pick a starting number between 1 and the k, q
- Sample the qth individual, then the (q+k)th and then the (q+2k)th and so on
Describe stratified random sampling
Population is divided into different categories
Different random samples taken from each strata
What are the non sampling errors?
- Selection bias
- Self-selection bias
- Interviewer effect
- Question effect
- Non-response bias
- Behavioural considerations
- Survey format
- Transferring findings
What are the reasons for replicating an experiment?
- Assess the amount of natural variation in the results
2. Increase precision
What is an experiment?
In an experiment, we try to discover whether a treatment or condition has an effect on individuals
What are experimental units known as if they are human?
Subjects
What are the two types of observational studies?
Prospective
Retrospective
What is a prospective study?
Looking forward
Choosing samples now, measure variables and follow up in future
What is a retrospective study?
Looking backwards
Examine previous exposure
What is the formula for the number of pairwise comparisons, if n is the number of different situations?
n(n-1)/2
