Find An Allergist / Immunologist | Pollen Counts | Journal of Allergy and Clinical Immunology | Annual Meeting   Site Map   Contact   Home    Search
Members Member Resources AAAAI News AAAAI eNews AAAAI Job Placement Center Promoting your Practice Order Public Education Materials Disease Management/ Ask the Expert Allergy, Asthma & Immunology Education and Research Trust (ART) Journal of Allergy and Clinical Immunology 2007 Accomplishments Annual Meeting Executive Staff Contacts Online Membership Directory AAAAI Organizational Structure » Members Contact AAAAI: Professional Home for the Practicing Allergist Grants and Awards Hot Topics Joint Council of Allergy, Asthma and Immunology (JCAAI) ST*AR Program AAAAI Assembly overview Allied Health Federation of the RSLAAI Societies International Military Allergy New Allergist/Immunologist Assembly Program Directors Organizational Chart Board of Directors Interest Sections Committees Fellows-in-Training Allied Health Professional Assembly Federation of RSLAAIS International Assembly Military Allergy Assembly New Allergist/Immunologist Assembly Program Directors Associates		AAAAI Quality of Life Resources Psychometrics A survey is evaluated by its psychometric properties. Psychometrics is the study of a scale's reliability, validity and sensitivity. Reliability refers to the extent to which the measure yields the same score each time it is administered, all other things being equal. Reliability is most often measured using Cronbach's alpha, which can range from 0 to 1. If a scale has an alpha value above 0.7 the scale is usually considered to be reliable. Validity is the degree to which the measure reflects what it is supposed to measure. Content validity is the extent to which the items on a survey measure the content under study. Sensitivity is the ability of a survey to detect a change when an intervention is given. Many of the most common surveys use a 1-5 Likert scale, which allows more sensitivity than simple yes/no surveys. Potential threats to the sensitivity of a scale include "floor" and "ceiling" effects. If patients are given a survey where 1 indicates a low QOL and 5 indicates a high QOL and they fill it out with all 5's before the intervention then we have a ceiling effect and this survey will not be able to detect any positive changes due to the intervention. Finally, in biostatistics the difference between statistical significance and clinical significance is often discussed. A frequently used statistic in HRQoL research is the paired sample t-test in which patients are tested before and after a medical intervention. A t-test is computed and we can evaluate if the treatment led to a statistically significant result. It is often assumed that a result which is statistically significant is also clinical significant. This assumption may not be true. There are situations in which a very large sample is used for a study and, even if the treatment effect is minimal, a large sample size practically guarantees statistical significance. Some researchers provide an index of what the minimal treatment effects should be when using their scales so that meaningful clinical conclusions can be drawn when analyzing HRQoL results. <back>
	© 1996-2008 · All Rights Reserved · American Academy of Allergy Asthma & Immunology