Reliability and Validity

Researchers tend to be peculiar with how they conduct their study and which study method they choose to implement on their subjects. When various tools are examined for measure in behavioral research, researchers tend to prefer high validity and reliability (Rosnow & Rosenthal, 2008). To understand this concept, one must comprehend the meaning and the difference between reliability and validity and how this applies to human services research. Without the full knowledge and comprehension of these two important criteria in determining how well measurements and particular research designs complete their functions, it will prove to be exceptionally challenging to display how the measurement completes its function or how stable it is.

Validity basically displays how well the measure or design does what it is designed and intended to do (Rosnow & Rosenthal, 2008). Validity is one of the most important criteria to determine how well measurements and particular research designs complete their functions (Rosnow & Rosenthal, 2008). Alongside validity is reliability within the criteria requirements and importance. Reliability implies stability or consistency (Rosnow & Rosenthal, 2008). Dependability may be implied. To state this plainly, if the type of measurement wishing to be applied is deemed unreliable, it is most likely because it is not valid. However, what should also be considered is that it is plausible for a measure to be reliable but not necessarily valid (Rosnow & Rosenthal, 2008).

Several types of reliability exist. Of the few, one of them is called Alternate-form reliability, which is defined as the amount of relatedness of altered forms of the test (Rosnow & Rosenthal, 2008). Another type of reliability is called Internal-consistency reliability. The meaning of this form is defined as the overall extent of significance of all items in an assessment or all raters in a judgment research (also known to be called reliability of components). Test-retest reliability is another form used in research design. Its definition can be identified as the degree of carnal stability (relatedness) of a scaling apparatus or test, or the characteristic it is created to calculate, from one administration to another; also known to be labeled as retest reliability (Rosnow & Rosenthal, 2008).

Validity also has its different types. One form is validity is called Content validity. This form is known to be the appropriate sampling of the applicable material or content that a test connotes to measure. External validity refers to the generalizability of an inferred fresh or new relationship over various people, settings, manipulations (or treatments), and research results (Rosnow & Rosenthal, 2008). Another form which contrasts the meaning of external validity is labeled as Internal validity. This form is characterized as the correctness of statements about whether one variable is the cause of a specific result, especially the capacity to cancel out plausible rival hypotheses. Face validity is something that is known as the degree to which an analysis or other forms of device “looks as if” it is calculating something pertinent (Rosnow & Rosenthal, 2008).

One type of data collection is identified as a survey method design. Survey research entails any measurement procedures that involve asking questions of participants. A survey can be anything from a short paper-fill-out feedback to an intensive one-on-one in-depth interview (Trochim, 2006). Surveys are roughly divided into questionnaires and interviews. Selecting the proper survey method that is best suited for the situation is important. Once the accurate survey method is chosen, constructing the survey itself is next in the process (Trochim, 2006). Surveys make it possible to collect data from large or small populations.

As with most aspects within research, strengths and weaknesses are inevitable with mostly everything. Surveys are relatively inexpensive, they are useful in describing the characteristics of large populations, can be conducted from remote locations by mail, e-mail or telephone (Trochim, 2006). Large amounts of responses are more valued in which it makes the results statistically significant even when analyzing multiple variables. Numerous questions can be asked of a topic giving flexibility to the analysis (Trochim, 2006). High reliability is usually easy to obtain with all participants being shown a stimulus that is standardized (Trochim, 2006). Surveys are more likely to be weak on validity and strong on reliability. Careful wording, format and content can increasingly reduce the participants’ own unreliability (Trochim, 2006).

Many various aspects are included into conducting and implementing an accurate research of measurement because there are so many obstacles and uncertainties that accompany the method. Researchers who have had an intense amount of education and experience still face difficulty with obtaining high validity and reliability into their research measurement. This is because various types of reliability and validity are present. This also includes countless ways of implementing and conducting data collection within the methods used and how it pertains to human services research or managerial research. Each has his or her own distinct type of analysis and how this affects certain types of human service behavioral research.

Reference
Barribeau, P. (2013). Commentary on Survey Research. Retrieved from http://writing.colostate.edu/guides/page.cfm?pageid=1421 Rosnow, R.L., & Rosenthal, R. (2008). Beginning Behavioral Research: A Conceptual Primer (6th ed.). Retrieved from The University of Phoenix eBook Collection database. Trochim, W.M.K. (2006). Research Methods Knowledge Base. Retrieved from http://www.socialresearchmethods.net/kb/survey.php