Methods II Flashcards
What is sampling
The process of selecting entities from a larger population, so that by studying the sample we might infer something about the population
What is simple random sampling?
select a ample of random entities from the population such that they have no pre-study correlation: e.g. based on IBAN, or phone number.
What is stratified random sampling?
Sampling such that every subgroup (of intrest) is represented and the possibility of biasing is reduced.
What is cluster random sampling?
Sampling of a population that is known to be a subgroup of a larger population: e.g. sample only people from a certain city or with a certain occupation.
what is convenience random sampling?
The sampling of entities of a population based on what is readily or easily (conveniently) available to the researchers. eg. friends, family, coworkers.
What does the sampling distribution (graph) tells us?
How much random error we should expect just from the sampling process.
Is this true?: “a random sample is always representative”
No, because a random sample can be anything, so also a set of just extreme cases.
A sample size of 30 is sufficient for a sampling distribution
False, n=30 is when sampling distribution comes close the gaussian distribution. If it is sufficient depends on the study.
“It must be possible for a hypotheses to be false”
True, if it is not falsifiable, there is no point of studying it.
What is meant with validity?
A systematic error in your study. often introduced by poor measurement equipment or ignorance of environmental influences by experimenter. often CAN BE eliminated if identified.
What is meant with reliability?
A random error (or noise) in the outcome of your studies. Unpredictable fluctuations that neither can be controlled or explained. Often it is possible to reduce the effect of this error.
“With statistical analysis we study the systematical error”
False, the systematical error is determined by the contours of the study itself. Sys.an. has to do with the random error of the data, and finding an explaination.
What is the description of a main effect of a response factor?
When direct changes in the response factor create a destinct difference in the response, without other variables. e.g. all longer conversations lead to more profit compared to shorter ones, indepent of email or phone communication.
What is the description of an interaction effect of a response factor?
When the effect of an response factor is dependent on an other factor. e.g. although longer conversations earn more profit; longer phones > longer emails while short phone < short email. Hence the communication type is a response factor with an interaction effect.
Determine what effects are present: (main effect of channel, main effect of length, interaction effect)
main effect of channel
Determine what effects are present: (main effect of channel, main effect of length, interaction effect)
Interaction effect