For Questions 1 - 7, please check the box next to the item that you think best completes the statement. These items are worth 3 points each.

1. Professor Anderson developed a NEW test to measure IQ (intelligence quotient). She claimed that someone scoring an IQ of 180 on her test could be considered twice as smart as someone scoring an IQ of 90, and that someone scoring an IQ of 90 would be three times smarter that someone scoring an IQ of 30. Anderson’s test treats IQ as a(n):

□    nominal variable
□    interval variable
□    ratio variable
□    ordinal variable
□    simple indicator variable

□    ensuring accuracy
□    ensuring precision
□    ensuring that your measure measures what you think it should measure
□    whether a particular measurement technique applied repeatedly to the same unit of observation would yield the same results each time
□    all of the above

□    describing a population that is too large to observe directly
□    descriptive, exploratory, and explanatory purposes
□    the measurement of attitudes prevalent in a larger population
□    studies where individual people are the units of observation
□    all of the above

4. Professor Southerland performed an experiment on learning by varying the mode of instruction. She did this for three different grade levels. Which of the following is (are) the independent variable(s)?

□    amount learned
□    grade level
□    mode of instruction
□    the second and third listed above
□    all of the above

5. Which of the following is (are) most likely to produce clear evidence of a causal relationship between the dependent and independent variables?

□    natural experiment
□    naturalistic experiment
□    true experiment
□    the first two only
□    all of the above

6. In a study to determine if apartment rental agents are discriminating on the basis of race, the researcher identified several pairs of adults where the members of each pair were closely matched in terms of their socio-economic characteristics, except for their race. He then sent each person in a matched pair to individual apartment complexes to see how they were treated. This is a good example of:

□    an experiment with ethical problems
□    a quasi-experiment
□    a naturalistic experiment
□    a natural experiment
□    an experiment with low reliability

7. Most social scientists would not accept the operationalization of Intelligence Quotient (IQ) through the use of foot size because such a measurement lacks:

□    precision
□    reliability
□    accuracy
□    validity
□    all of the above

Please indicate whether you think the following eight statements are true or false by checking the box next to your choice. These items are worth 3 points each.

8. A personal question like “What is your household’s annual income?” must always be asked toward the very end of a survey questionnaire.

9. All other factors held constant, the more heterogeneity there is within a universe for a specific variable of interest, the larger will be the required sample size.

10. “Time series” data is composed of two or more cross-sectional measurements of one or more variables.

11. Positive questions deal with “what is” and normative questions deal with “what should be.” Hence, public policy making is enhanced most when scientific methods are used to find the correct answer to normative-type questions.

13. In a typical SPSS data set (in the data view), the units of observation are represented in columns and the variables are represented in rows.

14. You can tell if a research study is an evaluation of a policy or program by the type of experiment used.

For the remaining questions on this exam, please constrain your answers to the space provided, and write legibly.

15. You have been asked by an environmental advocacy group to conduct a poll to determine whether citizens in the U.S. believe that global warming exists. What is the question you would pose to the respondent? List three factors which affect the required size of a random sample of respondents to be drawn from the universe of persons comprising the U.S. population. (12 points total)

16. Consider the following issue: Does the number of historic preservation advocacy groups within a U.S. state increase the proportion of potential sites in that state placed on the register of historic places? Develop an outline of a true experimental design, assuming that there are no constraints of time, money or your capacity to implement the design. (18 points total)

What attributes of the true experimental design would change if the question were to be examined using a natural experimental design? (3)

December 9, 2008, 4:46 pm
Being Paid to Lose Weight

Anyone who has ever tried to lose weight knows it’s tough to stay motivated. Researchers at the University of Pennsylvania wondered if cold, hard cash might do the trick.

The investigators studied how cash incentives influenced weight loss among 57 people who were obese but otherwise healthy. One group of dieters was given $3 a day plus additional matching funds for meeting weight loss goals, meaning they could earn up to an extra $168 during a four-week period. A second group was eligible to compete in a cash lottery for $10 to $100 a day during the study period if they achieved weight loss goals. A third control group simply attended monthly weigh-in sessions.

At the end of 16 weeks, the dieters in the lottery group had lost an average of 13.1 pounds each, while those in the matching-funds group had lost an average of 14 pounds each, compared to just 3.9 pounds in the control group, according to a report published today the Journal of the American Medical Association.
Five people in the lottery group and seven in the matching-funds group lost more than 20 pounds each during the study period. Only one person in the weigh-in group lost that much weight.

The downside was that some of these dieters gained back much of the weight once the cash prizes dried up. Still, seven months later they remained an average of six to nine pounds lighter.

The study offers lessons to businesses about the value of regular cash incentives to encourage healthy behaviors among workers, said Dr. Kevin G. Volpp, director of the Center for Health Incentives at the University of Pennsylvania School of Medicine.

“The goal is to design a reward system in a way to help people in the short term do what’s in their long-term best interest,” said Dr. Volpp, an internist and health economist. “A lot of insurers are starting to spend a lot of money on incentive programs to improve health. This shows that providing tangible rewards with a higher degree of frequency makes the use of these dollars more effective.’’

17. What was the major research question being investigated in this study? (2 points)

21. Identify at least one major strength of the study and explain why you think it’s a strength. (3 points)

22. Identify at least one major weakness of the study and explain why you think it’s a weakness. (3 points)

23. What is the difference between “internal validity” and “external validity” in an experiment and what, in your view, is the extent of internal and external validity of this study? (10 points)

2. Professor Smith gave an exam on Monday. On Wednesday Smith gave the same exam to the same class. He was assessing the exam’s... Circle ONE (2 points)

3. A researcher must decide in the process of operationalization... Circle ONE (2 points)

4. The independent (effect) variable must occur later in time than the dependent (cause) variable. (2 points)

5. If two variables are correlated with each other, there must be a causal relationship between them. (2 points)

6. List three factors which affect the required size of a random sample to be drawn from some universe of units of observation. (5 points)

7. In one sentence, what distinguishes a parametric statistical test from a non-parametric statistical test? (5 points)

8. In one sentence, state the difference between a variable that is a scale and a variable that is an index. (5 points)

9. In no more than two sentences, state the difference between cross-sectional data and longitudinal data. (5 points)

12. Select ONE of the following concepts. In ONE phrase/sentence, how would you conceptualize it? In ONE phrase/sentence, how would you operationalize the concept? Give two examples of your indicators and explain its level of measurement. (10 points)

13. Describe a study for which the analysis of existing data is the appropriate research method. What is the research question? What data source would you use? Specify ONE variable in the study for which you would gather data from that data source. Indicate at least one problem that you might encounter and how you would resolve it. (10 points)

14. There has been speculation over time that representation of defendants by public defenders versus privately retained attorneys affects the outcome of criminal trials in Delaware’s Superior Court. Develop a true experimental design, assuming that there are no constraints of time, money or your capacity to implement the design. What would be the research question? The unit of analysis? The dependent variable? Three independent variables? What attributes of the true experimental design would change if the question were to be examined using a natural experimental design? (20 points)

15. Discuss the strengths and weaknesses of the research study described in the article below. What was the research question? As part of your response, be sure to address any important reliability & validity issues. (20 points)

New York Times, December 9, 2003
Patterns: In Aging, Distress Takes Its Toll
By JOHN O'NEIL

People who get more upset by disturbing events are more likely to suffer the declines in memory and mental ability found in Alzheimer's disease, according to a study published today. The study tracked a group of priests, nuns and monks as part of a long-term examination of the aging process.

Dr. Robert S. Wilson of the Rush Alzheimer's Disease Center in Chicago, the study's lead author, said that earlier research had shown that chronic stress undermined the functioning of the part of the brain governing memory. He and his colleagues wondered if a lifetime of stress could make people more vulnerable to Alzheimer's.

They tested the idea on members of the Religious Orders Study, whose life experiences were more similar than the general population's. Eight hundred members, with an average age of 75, completed surveys in an effort to gauge what researchers called "distress proneness," how likely reactions to stress would result in gloom or anxiety, a trait also known as neuroticism. Various memory and other mental tasks were also measured.

The subjects were re-examined an average of five years later. In the meantime, Alzheimer's had been diagnosed in 140 of them. People who had scored the highest on the neuroticism test were twice as likely to have developed the condition as those who scored the lowest, whom Dr. Rush described as "secure, hardy and generally relaxed," even when factors like depression were accounted for.

Even among those who had not developed Alzheimer's, higher scores on the test were associated with faster rates of decline, particularly in memory.

1. (20 points) A nonprofit medical clinic in Philadelphia provides counseling and medicines to persons who are infected with the HIV virus. The primary aim of the organization is to improve the quality and supply of counseling services and medical treatments available to this population. The clinic realizes that it has never evaluated its program and comes to you for help. Develop an outline for an outcome evaluation of the program. For purposes of this question, assume that the services are being delivered as planned. For this outcome evaluation study, what would be appropriate: (1) research questions; (2) experimental design; (3) data to be gathered (specific data); (3) methodologies to use?

2. (20 points) Consider the following research question: Does the presence of national environmental protection agencies among nation-states affect the quality of the environment in those locations? Develop an outline of a true experimental design, assuming that there are no constraints of time or money. What would be the research question? The unit of analysis? The dependent and some of the independent variables? What attributes of the true experimental design would change if the question were to be examined using a natural experimental design?

3. (20 points) You’re a member of a local school board in a school district which is experiencing significant growth in its younger school-age population. The board must decide whether it should build a new elementary school or expand an existing one. While the costs associated with each option are important to consider, there are several other important factors. You suggest that one of these other important factors is the perspective of the public. You propose that a telephone survey be done to obtain this information, and before you know it, you find yourself the chair of a committee charged with designing and overseeing the execution of such a survey. Please prepare an outline of a report to the school board which describes your committee’s proposed approach for this survey, keeping in mind that none of your colleagues on the board have had formal training in research design and methodologies. Your report outline should answer the following questions:

4. (20 points) Draft five (5) questions that would be part of the survey described in Quesion 3 above. For each of these questions:

5. (20 points) Discuss the strengths and weaknesses of the research study described in the article below. If you feel you don’t have enough information about the study and how it was conduced, what would you like to know about the study that is not reported in this article and why. As part of your response, be sure to address any important reliability and validity issues. (20 points)

Please return this sheet along with your blue book(s) AND the
completed course evaluation questionnaire.

1. What are the main differences between nominal, ordinal, interval and ratio data? Give examples of each. Given these differences, what measures of central tendency are appropriate for each type of data? (10 points)

2. Make a distinction between primary and secondary data and state the advantages and disadvantages of each. (10 points)

3. A nonprofit organization rehabilitates houses in inner-city Chicago. The primary aim of the organization is to improve the quality and supply of housing available to low income people. The agency realizes that it has never evaluated its program and comes to you for help. Develop an outline for an outcome or summative evaluation of the program. What would be appropriate: (1) research questions; (2) research design; (3) data to be gathered (specific data); (3) methodologies to use? (20 points)

4. Consider the following research question: Does the knowledge of the owners of historic buildings regarding the buildings' historical significance increase the likelihood that the owners will preserve or restore those buildings? Develop an outline of a true experimental design, assuming that there are no constraints of time or money. What would be the research question? The unit of analysis? The dependent and some of the independent variables? What attributes of the true experimental design would change if the question were to be examined using a natural experimental design? (20 points)

5. Discuss the strengths and weaknesses of the research study described in the article below. As part of your response, be sure to address any important reliability & validity issues. (20 points)

Researchers link mice to inner-city asthma
BALTIMORE -- (AP) -- Common house mice may be a major contributor to asthma among inner-city children, according to a federally funded study.

Asthma affects approximately 15 million Americans, including 5 million children. An estimated 7 percent of children nationally have the disease, though researchers say it's twice as common in inner cities. In a study involving eight urban areas, Johns Hopkins University scientists discovered that 95 percent of tested homes had at least one room containing mouse allergen -- substances that cause allergic reactions, such as mouse urine or dander. Baltimore topped the charts at 100 percent.

The study, published in the December issue of the Journal of Allergy and Clinical Immunology, found that 18 percent of the children in the homes were allergic to mice and tended to have severe asthma. For years, researchers have known that cats, dogs, dust mites and cockroaches can cause allergies that trigger the wheezing and constricted air passages of asthma. ``While cockroach is the more important allergen, mouse is second in line,'' said Dr. Robert Wood, associate professor of pediatrics at Johns Hopkins and the study's lead investigator. Researchers used data from the National Cooperative Inner-City Asthma Study, completed in 1996, which covered 1,528 children ages 4 to 9 in Baltimore, Washington, the Bronx and Brooklyn in New York City, St. Louis, Chicago, Cleveland and Detroit.
Continued on reverse side

The children all were diagnosed with asthma and lived in neighborhoods where 30 percent or more of the households had incomes below the 1990 poverty level. Researchers analyzed dust samples from the homes of 608 of the children in the earlier study. The highest levels of mouse allergen were found in kitchens, followed by bedrooms and then living rooms. In all, 87 percent of the sampled rooms had detectable mouse urine or dander.

The study was funded by the National Institutes of Health, the Environmental Protection Agency and the Center for Indoor Air Research.

6. Policy Maker Interviews: You and a colleague are interested in studying the attitudes of representatives to international organizations regarding global warming. You decide you will interview the ambassadors to the United Nations from member countries to understand their views on global warming and their votes in the UN regarding the issue. Your research question is: What are the ambassadors' views on global warming and how do their views affect their votes in the United Nations General Assembly regarding this matter?

Your colleague has written up the first draft of the questions you will both be administering to policy makers. She asks for your comments on them. Make at least five suggestions for revision of the wording and ordering of the questions listed below. Rewrite the new questions in the appropriate order. If you feel that important questions have been omitted you may add them (but this is not required). Explain why you are suggesting each of the changes you have made, including why you have ordered the questions the way you have. (20 points)

Please return this sheet along with your blue book(s) AND the completed course evaluation questionnaire.