Posted: April 20th, 2022

PH 772: Epidemiologic Research II

Final Exam
PH 772: Epidemiologic Research II
Monroe College
Winter 2022

NAME __________________________________________________________

Q1. Cohort studies can be _____________ in nature:
A. Prospective
B. Retrospective
C. Ambispective
D. All of the above

Q2. Which of the following designs is best for testing a causal relationship?
A. Cross-sectional
B. Ecological
C. Case-control
D. Experimental trials
E. Case report/case series
Q3. Hypothesis are most likely to be included in what type of research studies?
A. Descriptive
B. Analytic
C. All of the above
D. None of the above
Q4. Conducting a descriptive analysis is usually the first step in the quest for causality.
A. True
B. False
C. None of the above
Q5. Which of the following will help to mitigate confounding?
A. Multiple regression model
B. Simple linear regression
C. Simple logistic regression
D. All of the above

Q6. To be considered an antagonistic interaction, the joint effect has to be higher than the effect expected by the sum their individual effects.
A. True
B. False
C. None of the above
Q7. The outcome variable in a multiple linear regression is best measured on which of these scales?
A. Categorical
B. Quantitative
C. Ordinal
D. All of the above
Q8. Logistic regression analysis can help us determine the increase in an interval dependent variable for each unit increase in the independent variable.
A. True
B. False
C. None of the above

Q9. R2 is the modified version of R that has been adjusted for the number of predictors in the multiple regression model.
A. True
B. False
C. None of the above

Q10. Like regression model, correlation can be used to express the functional relationship between X and Y.
A. True
B. False
C. None of the above

Q11. Slope predicts the change in Y per unit change in X, thus providing a measure of the effect of X on Y.
A. True
B. False
C. None of the above

Q12. You want to find the prevalence of e-cigarette smoking is a given community. Which of these study designs would you use to test the hypothesis?
A. Case-control
B. Cross-sectional
C. Cohort
D. Experimental trials

Q13. A study asks, “What is the frequency of fast-food consumption in urban adolescents?” What kind of research question is this?
A Descriptive
B. Association
C. Causal
D. None of the above

Q14. The degree to which a measure performs consistently from one time to another is known as:
A. Predictive validity
B. Concurrent validity
C. Discriminant validity
D. None of the above
Q15. You would like to explore the effects of gender, age, education and systolic blood pressure on COVID-19 infection (1=positive; 0=negative). What is the most appropriate regression model to use in this analysis, and why?

Q16. Let’s assume you conducted a statistical analysis designed to examine the association between race and diastolic blood pressure (DBP) in a small community in New York City.
a. State your research question.
b. State your hypotheses (both the null and the alternative).
c. State the dependent and the independent variables.
d. Use an appropriate statistic to test the hypothesized relationship; interpret your findings and draw your conclusion.

Q17. Using the Practicum Data, recode systolic blood pressure (SBP) into the following three categories, and assign values to the created categories:
1. 87 – 120
2. 121-140
3. 141-339.

Q18. Run the frequency distribution of the recoded variable and interpret the results.

———-

PH 772 Final Exam: Epidemiologic Research II

Monroe University

Winter of 2022

NAME

Q1. Cohort studies can be of the following types:

A. Predictive

B. A Look Back

C. Prospective

D. All of the aforementioned

Q2. Which of the following designs is most appropriate for testing a causal relationship?

A cross-sectional study

B. Environmental

C. Case-control study

D. Experimentation

E. Case study/case series

Q3. What types of research studies are most likely to include hypotheses?

A. Narrative

B. Analytic

C. All of the above

D. None of the above

Q4. Conducting a descriptive analysis is usually the first step in the quest for causality.

A. True

B. False

C. None of the above

Q5. Which of the following will help to mitigate confounding?

A. Multiple regression model

B

Order | Check Discount