1. What does poverty mean to you? Have you been using an absolute, relative, or objective standard when you think about poverty? Identify two examples of “poverty” and explain why they represent this concept. Compare your conceptualization with those of your classmates and what you find in a dictionary. Can you improve your conceptualization based on some feedback?
2. What questions would you ask to measure feelings of being “in” or “out” with regard to a group? Write five questions for an index and suggest response choices for each. How would you validate this measure using a construct validation approach? Can you think of a criterion validation procedure for your measure?
3. If you were given a questionnaire right now that asked you about your use of alcohol and illicit drugs in the past year, would you disclose the details fully? How do you think others would respond? What if the questionnaire was anonymous? What if there was a confidential ID number on the questionnaire so that the researcher could keep track of who responded? What criterion validation procedure would you suggest for assessing measurement validity?
4. The questions in Exhibit 4.17 are selected from my survey of shelter staff (Schutt and Fennell 1992). First, identify the level of measurement for each question. Then, rewrite each question so that it measures the same variable but at a different level. For example, you might change the question that measures seniority at the ratio level (in years, months, and days) to one that measures seniority at the ordinal level (in categories). Or you might change a variable measured at the ordinal level, such as highest grade in school completed, to one measured at the ratio level. For the variables measured at the nominal level, try to identify at least two underlying quantitative dimensions of variation, and write questions to measure variation along these dimensions. For example, you might change the question asking, “What is your current job title?” to two questions that ask about the pay in the respondent’s current job and the extent to which the job is satisfying.

           What are the advantages and disadvantages of phrasing each question at one level of measurement rather than another? Do you see any limitations on the types of questions for which levels of measurement can be changed?