Reliability: Internal Consistency By Lynn Woolever AED 615 October 23, 2006
Internal Consistency Reliability refers to the consistency of scores obtained in an experiment. Specifically, the internal consistency method refers to the consistency of scores using only a single administration of an instrument.
Types of Internal Consistency There are three major types of internal consistency testing: Split-half Procedure Kuder-Richardson Approach Alpha Coefficient
Split-Half Coefficient In this procedure, the test is split in half and each half is scored separately– usually odd items versus even items. A coefficient is then calculated to determine if the two halves showed the same results. A coefficient is the degree to which the two halves of the test have the same results.
Split-half coefficient Calculated using the Spearman-Brown prophecy formula Reliability of = 2 x coefficient scores on total test 1 + coefficient
Kuder-Richardson Approach Most frequently employed method Reliability with this test should be.70 or higher The number of items, the mean, and the standard deviation are needed Typically use K 20 or K 21
Kuder-Richardson Approach K = number items, M = mean, SD = standard deviation K 21 Reliability = K 1 – M(K – M) coefficient K-1 K(SD 2 )
Alpha Coefficient This used to calculate the reliability of items that are not scored right or wrong such as an essay test where more than one answer is possible. Form of the K 20
Value of reliability and validity If the research survey is either not valid or reliable then the research won’t be valid or reliable. Without valid or reliable research credibility of the study is in question.
Use in Quantitative Research Gives reliability and validity numbers to support the quantitative research and thereby increase credibility of the study.
Use in Qualitative Research Because these studies often describe situations, instead of reliability and validity, researchers rely on honesty, believability, expertise, and integrity to prove the credibility of the study.
Journal Article One JAE article that uses Reliability and Validity is: Statistical Procedures Employed in the Journal of Agricultural Education. http://pubs.aged.tamu.edu/jae/pdf/vol31/ 31-04-46.pdf
Journal Article In the “Statistical Procedures Employed in the Journal Of Agricultural Education” article, internal reliability was used to test the reliability of different types of sources used in the Journal of Agricultural Education.
Work Cited Bowen, B., Rollins, T., Baggett, C. & Miller, J. (1990). Statistical Procedures Employed in the Journal of Agricultural Education. Journal of Agricultural Education, 31(4), 46-51.