Please enable JavaScript.
Coggle requires JavaScript to display documents.
SAMPLING (CHAPTER 7) (Important terms (Census: the collection and analysis…
SAMPLING (CHAPTER 7)
Important terms
Census: the collection and analysis of data from every possible case or group member (impossible due to restrictions)
-
-
Sample: sub group or part of a larger population (the complete set of cases from which the sample is taken)
Sampling size
the larger your size, the less likely that errors in generalzations about target populations are reduced
be aware of: the confidence of the researcher, the margin of error that the researcher can tolerate, the types of analyses the researcher is going to undertake (how many subgroups), the size of the target population
Central limit theorem: the larger the absolute size of the sample, the more closely its distribution will be to the normal distribution. --> when the researcher is doing statistical analyses, it is likely that he is coming to conclusions from these analyses about the target population (statistical inference)
Law of large numbers: samples of larger size arem ore likely to be representative of the population.
Sampling technique
-
Non-probability samples:
Quota: non random, used for structured interview being part of survey strategy. The sample will represent certain traits of the population as the variation in the sample is the same as in the population (type of stratified but then non-random) --> divide in (exclusive) subgroups (employees by type of educational degree), figure out the proportion (employees with a economics degree might be 1 out of 4), choose the size (if you are sampling 1000 people you might have a quota sample of 10), choose participants but use the quota (so 25% has to be those with a economics degree),
Purposive: use the judgement of reseach to select cases that will be able to answer the research questions (only occurs in small samples, and if you want to select informative samples (grounded theory), not representative for the target population but is only dependent on the research question
Critical case: selects on the basis that they can make a point dramatically or because they are important. (understand what is happening, so you can make logical generalizations)
Typical case: provides an illustrative profile using a representative case. You can illustrate what is typical.
Extreme case: focuses on unusual and special cases. You will learn how to answer the question most efficiently. --> NOT representative, and barely used
Theoretical: associated with grounded theory and analytic induction. the process of collecting, coding and analyzing data in a simultaneous manner in order to generate a theory. You will start with where you will collect the sample, not what.
Homogeneous case: one subgroup with similair traits, (in dept focus and discover minor differences)
Hetereogeneous (max. variation): uses researcher's judgement to choose the participants with enough diverse traits to provide the max. variation. Patterns represent key themes and allows the researcher to collect diversity.
Volunteer
Snowball: participants are not chosen, but they volunteered. Will be used when there are difficulties in identifying members from the target population. The most part is the initial part since you need to contact with one or two cases. The participants will then choose other potential respondents. This is not representative as people can choose other participants based on similar traits.
Self-Selection: the researcher allows each case to identify their desire to take part. You publish the need for cases through advertisement and then collect data from those who respond.
Haphazard (convenience/availability): one selects cases that are the easiest to obtain for the sample.
This will be used when there is no sampling frame. You will use a selection of techniques in which the chance or probability of each case being selected is not known. (subjective judgment, exploratory). Also there are no rules for the sample size. The clear focus lies on the logical relationship between your sample selection technique and the purpose and focal point of the research. Generalizations are being made to theory and not about the target population
Sampling technique: methods that reduce the amount of data the researcher needs to collect (alternative for census)
Advantage: can be more accurate as the cases are smaller so you have more time on designing and controlling the way of collecting data and the information you gather is more detailed
-
Sampling frame
the complete list of all the cases in the population from which a probability sample is drawn. If there is no suitable list then you should make your own.
-
-
-
To which extent can one generalize. If the sampling frame is all employees, then the researcher can only generalize to the employees
-
Data saturation
when the extra data that you have collected only provides little new info or new themes then you have reached data saturation
If the researcher collected data from a secondary source wihtin an organization, the response should then be 100% as the organization has already granted access