Stats: Statistical Reasoning Probability and Nonprobability Sampling Methods in Studies

independent variable

a variable that influences another varibale, also the variable that the researcher controls or manipulates to affect the dependent variable (X)

dependent variable

an outcome vairbale that is affected or influenced by the indepentent variable (Y)

Research question

what the research team is trying to answer through the study

null hypothesis

The hypothesis that suggests there will be no statistically significant efeect on the variables being studied

• researchers are always trying to prove it wrong

alternate hypothesis

The hypothesis that observations from a sample are influenced by a nonrandom element, the hypothesis the researcher is interested in

simple random sampling

• most commonly used

• this sampling occurs through a random selection of members from the sampling frame (everyone has the equal chance to be randomly chosen)

• mostly computer generated

stratified random sampling

is used when the researcher knows some of the variables within a population that will affect the representativeness of the sample

• ex: age gender ethnicity medical diagnosis

• ex: like need a sample with all grade levels so seperate all particpanrs into groups and then randomly pick from each group

what situations is cluster (complex) sampling used in

use this form of sampling is used in 2 situations

• when the time and travel necessary to use simple random sampling would be prohibitive

• when the specific elements of a population are unknown making it impossible to develop the sampling frame

how to conduct cluster (complex sampling)

to conduct cluster sampling a list of all of the states, cities, institutions or organizations associated with the elements can be obtained

• basically group out subjects and pick from them random subjects

systemic sampling

requires an ordered list of all the members of the population

• Individuals are selected through a process that accepts every

  kth member on the list using a randomly selected starting point

• Ex: If the population has 1000 potential participants and a sample

  size of 100 is desired then k = 1000÷100 = 10 – every 10th person is invited to participate in the study

Convenience sampling

Most frequently used

• Enroll subjects who are accessible and available to participate in the study

• Subjects are enrolled until the target sample size is obtained

• Does not allow for the opportunity to control for sampling errors or biases

What can quota sampling be used to mimic

the known characteristics of the population

quota sampling

used to ensure adequate representation of all types of subjects who are likely to be underrepresented (women, elderly, minorities)

why is quota sapleling an improvement over convenience sampling

because it decreases sampling error ir bias

what is quota sampling used in conjunction with

convenience sampling to help ensure the inclusion of the identified subject

purposive sampling

occurs when the researcher consciously selects subjects, elements, events ir incidents to include in the study

• completely opposite of random sampling, the subjects are truly handpicked

• This sampling method can be a good way to explore new areas of study

why has purposive sampling been critcized

because the researcher’s judgements in the selection of the cases cannot be evaluated or understood

network/ snowballing sampling

makes use of social networks and the fact that friends often have common characterisitcs

• The researcher identifies a few subjects who meet the sampling criteria and then asks them to assist in recruiting others with similar characteristics

• Useful for gathering samples that are difficult to obtain, have not been

  previously identified for study or can help the researcher explore a particular

  area of interest

• Sampling errors or biases are inherent with this method

what are the 4 elements to a power analysis

• Alpha or level of significance (usually this is 0.05)

• Standard power (usually this is 0.80 or 80%)

• Effect size (strength of relationships) (how big are the differences→ if the effect size is going to be really big dont need a big number of ppl

• Sample size