9 What is reliability? Reliability is concerned with the consistency of the measurement method. 10 What are the three aspects of reliability testing ? a. stability
b. equivalence c. homogeneity 11 What is stability? A type of reliability testing that focuses on the consistency of results when a test is repeated. It is also called test-retest reliability. It is expressed as an "r" value with higher numbers = greater test-retest reliability. 12 What is equivalence? A type of reliability which involves the comparison of two versions of the same pencil and paper instrument or two observers who are observing (or grading) the same event. a. Two versions of the same test = alternate forms reliability b. Two judges rating the same person = inter-rater reliability c. It is expressed as an "r" value with values above .8 = greater IRR or alternate forms reliability. 13 What is homogeneity? A type of reliability testing that is used with pencil and paper testing. It addresses the correlation (or relationship) of each question on the test to other questions on the test. It basically asks...are the questions all asking about the same basic constructs? This is also known as internal consistency. 14 What statistical test is used to check a pencil and paper test for homogeneity or internal consistency? A Cronbach's alpha or a Kuder-Richardson's 20. Both of these are expressed as an "r" value with values above .8 = high internal consistency. Note: internal consistency is checked every time a pencil and paper test is used. If your article for critique used a questionnaire of some sort, then they should report this number for both the questionnaire's development or past use AND the current sample. 15 What is validity? How well does the instrument reflect the abstract concept being measured? Hint: validity exists on a spectrum. It is not an all or nothing phenomenon. We try to determine the degree of validity an instrument or questionnaire has for a specific sample or situation. 16 When looking at a measure or questionnaire for validity, we typically descrive this as content validity. Content validity can be broken down into two subtypes. What are they? a. Content and predictive validity. 17 What is content validity? The extent to which the measurement / questionnaire / scale includes all of the major elements or items relevant to the construct being measured. How do you prove this? a. 1. do the items in the scale reflect the description from the ROL? b. 2. how do content experts rate the items on the scale? c. 3. how might potential subjects respond to items on the scale? 18 What is evidence of validity from contrasting groups? Identifying groups that are expected (or known) to have contrasting scores on a questionnaire. a. Example: administering a questionnaire about depression to two groups of people, one group with known depression, one group without, and comparing their scores.
You've reached the end of your free preview.
Want to read all 30 pages?
- Fall '16
- Denise Cauble
- Nursing, Qualitative Research, researcher