Stat 1-3 Flashcards
Simple Random Sampling
Simple Random sampling is when individuals are randomly selected from the total population, ex. trying to find a sample for 4 workers in store with 4 departments through simple random sampling would have you take the names of all workers from all departments and then draw 4 random names, (note this would not ensure that the workers are from different departments you could draw 4 names from the same department)
Stratified random sampling
Stratified random sampling is when you split a population into categories and make groups out of people who fit into the same category called stratum, (all the combined stratum are called the strata), then use simple random sampling to select individuals from each group, therefore ensuring that each category is represented in the sample. EX taking a stratified random sample of 4 workers from a store with 4 departments, would envolve first splitting each department as its own stratum and then creating your sample by selecting one individual from each stratum using simple random sampling.
Systematic random Sampling
Systematic random sampling is when the members of a population are listed and choosen at a numerical interval to be part of the sample. If the individuals position on the list is not believed to affect their responses it is a representative sample. Ex taking a systematic random sample of 4 workers from a store with 4 departments would involve listing all workers in the store and choosing every tenth worker to be part of the sample
Cluster random sampling
Cluster random sampling is when a population is split into groups. Cluster random sampling is unbiased if the clusters are thought to fairly represent the population, and one of the groups is choosen randomly. EX if a store wants a sample of 10 people and has 4 departments each made of 10 people and is looking for a sample of 10 workers cluster random sampling would involve randomly choosing one of the departments. If the individuals in that department were thought to represent the entire population of workers in respect to the criteria then it would be considered unbiased
Selection bias
Occurs when a portion of society is under or not represented at all in the sample.
Ex if you take a sample via email you will exclude the portion of the population that can not access/dosent have email
Voluntary response bias
When the only means for selecting the sample is through volunteers, may only get people specifically interested in the issue, and not the attitude of the overall population
Response bias
When the format of the survey discourages participants from choosing a certain answere
ex. asks about illegal activity
random allocation
allows us to make a causal inference
means that individuals were randomly placed in each group not which group they wanted
random sampling
allows us to make a population inference
define observational studies and list the two types
observational studies are when reserachers do not attempt affect variables and instead look for relationships and trends
retrospective studies: uses data from the past, good if it is for a rare occurance but there is a greater possibility for observational error
prospective studies: when reachers continue monitoring a variable for change for a certain ammount of years good in that they have a greater ability to spot lurking variables and a lesser chance for observational error, but can be more expensive
What are the two types of data
Categorical and quantitative
Categorical data when data is not numbers/if it is numbers the numbers are being used to represent individuals/objects. The two types of categorical data are nominal and ordinal. Nominal data is when the order doesnt matter, ie gender of participants. Ordinal data is when the categories can be ranked in the same list each time, ie letter grades.
What is the parameter
P like population, parameter is when a characteristic can be generalized to a population
What is a statistic
Statistic like sample is when a representation of a characteristic in a sample
