1. In the opening section here, what is meant by “generalization.” Why do you want to do this?
Answer: Generalization refers to the idea of creating a basis for all human beings, included inside and outside an area of study. Researchers should not just focus on the people in the study; they should be able to take these findings and data and spread it to other people in the population. A person wants to generalize about the people outside of their focused area of study. Generalization can create an assumption for the rest of the population of a certain group. For example, a survey can be given to a portion of the residential population of Mount Mercy; from that information, a researcher could make a generalization about the entire residential student population at Mount Mercy.
2. When wouldn’t you be concerned about “generalizing”?
Answer: A researcher would not be concerned about “generalizing”, especially when looking at an organization, or sampling an organization. For example, a researcher-surveying people of Four Oaks, such as the clients For Oaks serves, could not be generalized because most of the clients come from different backgrounds; each situation at Four Oaks is different from client to client. Generalization should not be used in agencies, because the population is completely diverse.
Population: “The group a person wishes to generalize”; For example, a researcher would use Mount Mercy University as an institution as a potential population to generalize to find certain data.
Theoretical population: “The population a researcher would like to generalize”. For example, a researcher would take into consideration the student population on a certain topic for data collection.
Accessible population: “The population that is most accessible to the researcher.” In the example with Mount Mercy University, researchers have an easier time reaching students because they make up a large part of the stake-holding population of Mount Mercy.
Study population: “The population that you can have access too.” For example, a researcher might look for a list of email addresses for Mount Mercy students to use as a tool to send out an electronic survey.
Sampling Frame: “The listing of the accessible population from which you’ll draw your sample.” For example, a researcher might want to take a list of residential students that live on campus as a sampling frame. Researchers can then sample people from inside this category
Sample: The group of people whom you select to be in your study; those who respond to the study. For example, a sample for a researcher might consist of 75 responses from the residential population.
4. Describe the basic “sampling model”.
Answer: The basic sampling model found in the external validity portion of the website talks about taking a sample from a population, using the results of the sample, and then reflecting the results to represent the entire population of the group. The basic sampling model does not work in most cases because of the reflection of the results placed on the whole population from which the sample was taken from in the research. For example, if a researcher only received ten percent response from the males in the residential population; the data cannot relate to the men out of the general population. The basic sampling model does not properly reflect the general population.
5. What’s the difference?
Answer: The difference between probability and nonprobability sampling is the idea of random selection in the distribution and return process. Probability sampling really focuses attention on the theory of random selection in order to get the most results out of the surveys put out in a certain sample of the population. Nonprobability does not use the idea of random selection in the distribution of surveys to samples. Nonprobability sampling results should not reflect the entire population through the results of the survey. If a researcher wants the sample to reflect the population as much as possible, the researcher should focus more on probability sampling than nonprobability sampling.
6. What’s a “stratified random sample”?
Answer: A stratified random sample is a type of sampling where the researcher divides the population into certain, “homogeneous” groups that have similarities, and then takes a sample out of each group. The idea of stratified random sampling is to allow a researcher to get the best representation of the entire population, based off of the samples from the subgroups in your chosen population. For example, when researching students at Mount Mercy University, a researcher must look into subgroups of the student population, which are commuter students, residential students, part-time and full-time, traditional and non-traditional students. When a researcher takes random samples from each of these subgroups, the data can better reflect the general student population as a whole with the collaboration of results from the different sub groups.
Sampling Distribution: “The distribution of an infinite number of samples of the same size as the sample in your study.” In other words, a sampling distribution allows for a collection of data that can be used on a graph to show the number of responses a researcher will get compared to the total number of surveys given out. For example, if a group of 20 students was given a survey and only 5 of them gave responses, you could concur that only 25% of the population will give responses; that means out of a group of 100, only 25 would give responses back.
Sampling Error: “The standard error in sampling contexts.” In other words, this allows for some space to account for small statistical errors in a survey response. The sampling error reflects on the size of the sampling size; if the sample size is large, the sampling error will be smaller. Many different surveys place a sampling error in a visual graph of the results; there might be a 1.5% sampling error in which a number can be either 1.5 % less than reported or 1.5% more than reported.