If the subjects complete the scale in a similar way from one time to the next

# If the subjects complete the scale in a similar way

This preview shows page 5 - 6 out of 10 pages.

items similarly if they are given the same depression scale 2 days apart. If the subjects complete the scale in a similar way from one time to the next, this indicates test-retest reliability. 3.The Cronbach's alpha coeﬃcient (α or r ) is calculated to determine the homogeneity or internal consistency of multi-item scales with interval or ratio level of data in a study. This α or r value should be determined for every multi-item scale and its subscales used in a study, since a scale must be reliable for a particular study population in order to obtain useful data for analysis in a study. The Kuder-Richardson formula ( KR 20 ) is calculated to determine the internal consistency of instruments with data measured at the nominal level. Researchers also need to discuss the reliability of the scales used in their study based on previous research ( Grove et al., 2013 ). 4.A reliability coeﬃcient of 0.80 indicates the scale has 0.80 reliability or consistency in the measurement of a variable and 0.36 random error ( Grove et al., 2013 ). The calculation is as follows: 1.00 − (0.80 × 0.80) = 1.00 − 0.64 = 0.36. 5.If the Cronbach's alpha coeﬃcient is α or r = 0.55, then the scale has 0.55 reliability or internal consistency and a 0.697 potential for measurement error. The error calculation is: 1.00 − (0.55 × 0.55) = 1.00 − 0.303 = 0.697. This high potential for error means that researchers should question the quality of the data collected and probably not trust the results obtained using this instrument. If an instrument or scale is not reliable, it is not valid since a scale that does not consistently measure a variable is not valid. 6.Precision is the degree of consistency or reproducibility of measurements obtained with physiological instruments or devices. Researchers need to document their physiologic measure is from a quality manufacturer, implemented consistently with a protocol, and recalibrated according to manufacturers' directions. Higher levels of precision (0.90 to 0.99) are important for physiological measures to ensure that critical physiological functions are consistently measured, such as oxygen saturation. 7. Researchers reporting appropriate reading levels or readability scores for their measurement methods in a study enhance the reliability and validity of the instruments. Subjects must be able to read and understand the items on an instrument in order to complete it consistently and accurately. 8.Stability reliability was determined for the scales in this study by using test-retest reliability. The data collectors asked the children to identify their level of pain with each scale twice. If the scale scores matched on retest, a “1” was assigned, and a “0” was assigned if the scores did not match. The OUCHER scale had the highest test-retest reliability at 70%, and the VAS had the lowest reliability at 45%. Both the OUCHER and FACES (reliability = 67%) scales had strong test-retest reliability, and the data collected demonstrated consistency with repeated measures in this population. Since the VAS scale had the highest random error, the usefulness of this scale for this population requires additional investigation. The readability of the VAS for these children

#### You've reached the end of your free preview.

Want to read all 10 pages?

• Winter '16
• Cronbach