Test-retestreliability and inter-rater reliability
Boththe inter-rater and the test-retest are used to assess the externalconsistency. The two types of reliability differ in that thetest-retest approach maintains the reliability of the findings byfacilitating the performance of a given test in different periods,but using the same variables and the same raters (Holmeful, Aarts,Hoare & Sundholm, 2009). The inter-rater approach, on the otherhand, assesses the reliability of the same test, but using differentjudges. The inter-rater approach is based on the notion that raterswill disagree on certain issues, but the level of reliability oftheir findings can be easily determined by the level of consistencyin their results.
Constructand predictive validity
Bothconstruct and predictive validity indicates the degree of accuracy ofa given conclusion or measurement with respect to what was beingmeasured in a given test. Construct validity refers to the degree towhich a given construct actually measures the assertions that a giventheory claim. The evidence of validity involves the analysis of aninternal structure of a given test, which includes the assessment ofthe relationship between responses given to various test items(Brown, 2000). Predictive validity, on the other hand, indicates thedegree to which the score of a given assessment predicts thepossibility of a future performance. Due to its predictive nature,predictive validity is achieved by establishing a correlation thatexists between scores of a given test and other measures that thetest in question is intended to predict.
Internalversus external validity
Externalvalidity focuses on the degree to which the findings of a given testcan be generalized to a larger population. Internal validity, on theother hand, focuses on the degree to which confounding variables wereeliminated during the research or a test (Callans, 2012).
Itis possible for a given test to be reliable without necessarily beingvalid. This is because a given test can give consistent findings,which makes it reliable, but it only be considered to be valid if itis measuring what it was intended to measure (Callans, 2012).
Inconclusion, validity and reliability ad critical aspects consideredwhen determining whether a given research can be used to inform agiven decision of make viable conclusions. Reliability mainly focuseson consistency of the findings while validity focus on the accuracyof and the achievement of predetermined objective of a given test.
Brown,D. (2000). What is construct validity? Shiken,4 (2), 1-12.
Callans,M. (2012). Reliability and validity in testing: What do they mean?Wonderlic,Inc.Retrieved December 15, 2015, fromhttp://blog.wonderlic.com/reliability-and-validity-in-testing-what-do-they-mean
Holmeful,M., Aarts, P., Hoare, B. & Sundholm, L. (2009). Test-retest andalternate forms reliability of the assisting hand assessment. Journalof Rehabilitation Medicine,41, 886-891.