Essay About Questionnaire Reliability And Questionnaire Design

Ie 408 Hw 4
(Total: 1+1+1+1+2+2+2=10)(for last 4 questions, not use right method and conclusion wrong, -2; not use the right method, but conclusion right,-0.5; calculate wrong,-0.2; conclusion wrong / no conclusion, -0.1)14-2. What is the “horn effect” in questionnaire design? How does it differ from the “halo effect?”The horn effect is when a respondent perceives the situation so negatively that they rate all attributes poorly, even if some areas are not as bad, and therefore don’t deserve such low ratings.The halo effect is just the opposite. It is where the situation or person being rated is considered so positively that all responses are positively biased. Both of these effects lead to unreliable questionnaire results.14-5. What does the term “respondents demographics” mean? Provide an example.The responders demographics is a term that refers to background of the individual completing the survey or experiment. Often that information is useful as it may denote a potential attitude of the responder. It is also usually a covariant of the persons answers. For example, the age of the responder may affect how he or she views the desirability of a sports car. Demographics are important to determine whether perceptions differ among the age, gender, area of the country, or people with different experiences.14-10. What is questionnaire reliability? How can it be measured? And what is its significance?Questionnaire reliability is a measure or an indicator of how repeatable the results from the questionnaire are. The simplest measure of reliability is developed by asking the same question more than once. In a questionnaire if questions are repeated, a comparison can be made between the first half of the questionnaire questions paired with the second half, within each respondent.One other common measure of reliability is test-retest reliability, and is often found by administering equal forms of the same test to people and then computing the correlation coefficients within person between the two tests. The correlation should be high between answers by the same person. Another common method to test reliability is when half of the respondents answer one form of the questionnaire, while the other half of the respondents answer a second comparable form of the questionnaire. Then the two groups can be compared to determine if the two surveys prompt consistent responses. Reliability is especially likely to be a significant concern when trying to develop a simplified or shortened survey instrument, but is always an important issue. Simply put, unreliable or unrepeatable results cannot be relied upon.