discussion post

Using the assigned readings as a guide, answer the following question:  How can we take a valid sample in an educational setting?  The thread must use current APA formatting. 

Please review the Discussion Assignment InstructionsPreview the document and the Discussion Grading RubricPreview the document prior to posting.  You may also click the three dots in the upper corner to Show Rubric.
See attachments and respond to the following:
There are several aspects involved in taking a valid sample for educational research including the validity and reliability of the research instrument and methods, as well as the generalizability of the research findings to the population in question. As stated by Check and Schutt (2012), measurement validity refers to the extent to which measures indicate what they are intended to measure (p. 23).

There are several types of validity including 1. face validity, which is where the meaning of the concept is clearly related to what is being measured, 2. content validity, which ensures that the entire range of the concepts meaning is covered, 3. criterion validity, where scores from one measure can be closely compared to scores from an already validated measure, 4. construct validity, where the measure is closely related a variety of measures within a particular theory, and 5. convergent validity, which is similar to measure triangulation, where one measure of a concept is validated by being associated with different measures of the same concept (Check & Schutt, 2012). Construct validity is commonly used in validating research measures for educational investigations when there is no existing criterion to use for validation purposes (Check & Schutt, 2012). 

Reliability can also ensure that a research is valid. Reliability is when the research measures ensure that the scores are consistent and is a prerequisite for research to be considered valid (Check & Schutt, 2012). However, even though measures may be considered reliable does not necessarily mean they are valid. There are several way for researchers to ensure reliability including test-retest, interitem reliability, alternate-forms reliability, and interobserver reliability. Although it can be difficult to find methods that are both reliable and valid, the authors suggest being diligent about finding an appropriate measure to ones study and not selecting the first measure that one comes across in order to ensure the study can be used as valid research (Check & Schutt, 2012).

Proper sampling is an additional research method to ensure that data is valid and generalizable to the population that the researcher has selected for investigation. Researchers should first clearly define the population they wish to study and the elements within the population that will be investigated (Check & Schutt, 2012). The generalizability of a sample to the entire population will be determined by the quality of sampling methods and the measure of sampling error. Check and Schutt (2012) state that one way to ensure lower rates of sampling error is to increase the number of elements being sampled overall. The larger the sample size, the greater chance of generalizability to the population. There are four types of random sampling that one can use to lower the rate of sampling error including, simple random sampling, systematic random sampling, cluster sampling, and stratified random sampling (Check & Schutt, 2012). Other methods used mostly for qualitative research are non-probability sampling methods including availability sampling, quota sampling, purposive sampling, and snowball sampling. These methods are not as generalizable, however may be used when random sampling methods are not possible (Check & Schutt, 2012).

Check, J. W., & Schutt, R. K. (2012). Research methods in education. Sage Publishing.