Bshs 382 - Reliability and Validity

Reliability and Validity

BSHS/382

Reliability and Validity

Reliability and validity are important parts of qualitative research. This paper will define and describe reliability and validity as well as provide examples as they apply to human services management research. Data collection method and data collection instrument examples will also be provided in human services research as well as managerial research.

Reliability

There are several types of reliability. Below shows the different types and their definition:

* Inter-rater-: Different people, same test.

* Test-retest: Same people, different times.

* Parallel-forms: Different people, same time, different test.

* Internal consistency: Different questions, same construct (Changing Minds, 2012).

Inter-rater reliability evaluates different people while using the same test methods. An

example of this type of reliability would be if you asked a group of people to categorize pictures of fruits into citrus and non-citrus. To reach a reliable result both groups would have to categorize the pictures in the same way.

Test-retest can be affected by the subject's response to the test on any given day. The time of day, mood of the subject and numerous other variables can affect the outcome of the testing. An example of this type of reliability would be the SAT test. The test is conducted with classrooms of students every year. The questions asked help the researcher assess the validity of the test given over several years.

Parallel-forms reliability uses different questions for two separate tests at the same time. An example of this type of reliability would be gathering questions that would be split in half and asked of two randomly-selected target groups.

Internal consistency reliability is evaluating questions between individuals for comparison for consistency in appropriate results. An example of this type of reliability would be to ask one individual a question regarding a certain construct and then measuring that individual's answer against a second individuals answer to a different question with the same construct.

Validity

"Validity is the extent to which a test measures what it claims to measure" (Trochim, 2008). There are several types of validity to include content validity, criterion-related validity, construct validity.

Content validity is when the items represent the entire range of possible items. An example of this type of validity would be when a teacher