Exam 1

Which of the following is an industry standard software (i.e., typical software used

in the public sector) for public health research?

SPSS

Which of the following statements about the philosophy of science is FALSE?

A. Everyone, including scientists, are susceptible to logical fallacies, including atomistic & ecological fallacies

B. Scientific inquiry often rely on the process of empiricism, which involves induction & deduction.

C. Theories and empirical evidence may never overlap with one another

D. Falsification and demarcation criterion are the reasons why we try to “reject the null hypothesis” during analysis

Correct answer is C

The following diagram represents the nine steps of rigorous research. Where in the diagram does “identify a priority population” belong?

A. Step 1

B. Step 6

C. Step 8

D. Step 9

Correct answer is A Step 1

Quantitative research focuses on numbers, while qualitative research focuses on words.

True

Observational research is the most common type of research reported in public health

True

Which of the following is NOT an observational research design?

A. RCT

B. Longitudinal

C. Case-control

D. Multi-year cross-sectional

Correct answer is A

Cross-sectional research is very labor and cost intensive, with high concern for participant drop-out

False

Both cross-sectional and longitudinal designs take time flow into consideration

False

Case-control studies have retrospective design, and are useful for studying rare diseases

True

Experimental research designs eliminate threats to external validity, as they allow easy manipulation of variables

False

Which of the following is NOT a true experimental design?

A. Post test control group design

B. Pre test-post-test control group design

C. Interrupted time series design

D. Matched pair groups design

Correct answer is C

The following diagram is a representation of

A. Interrupted time series design

B. Matched pair groups design

C. Nonequivalent control groups design

D. Post-test control groups design

Correct answer is B

True experimental designs are more feasible for public health research, as they may use observational data

False

Interrupted time series design (ITSD) is the best quasi-experimental design for evaluating long-term effect of intervention

True

Nonequivalent Control Group Design (NCGD) posttest only design is also known as a within-subjects design

False

What would you do if you're in charge of testing the safety and efficacy of a new COVID-19 vaccine for the very first time?

a. Pretest-posttest control group design (typical RCT) for 30 people

b. Interrupted time series design (ITSD) for 2,000 people

c. Any quasi-experimental design for 5 people

d. Observational, cross-sectional design for 10,000 people

Correct answer is A

A Randomized Controlled Trial (RCT) is an example of exploratory research

a. True

b. False

False

Exploratory research often rely on observational research to set the stage for more

definitive research

a. True

b. False

True

Longitudinal research does NOT take time change into consideration, which limits its ability for causal inference

a. True

b. False

False

Cohort Sequential Design is a variant of the case-control design, and it's where persons serve as their own controls

a. True

b. False

False

Studies that use representative data generally have high external validity.

a. True

b. False

True

Which of the following words best describe what sampling is all about?

a. Representativeness

b. Size

c. Accuracy

d. Randomness

Correct answer is A

Which of the following are the three main components considered when making sampling decisions?

a. Sampling frame, sampling element, and sampling technique

b. Sample size, population mean, and confidence interval

c. Randomization, stratification, and matching

d. Data collection method, response rate, and measurement tool

Correct answer is A

All sampling methods are equally effective, so there’s no need to consider their strengths and weaknesses.

a. True

b. False

False

Which of the following is NOT a probability sampling method?

a. simple random

b. systematic

c. convenience

d. cluster

Correct Answer is C

Snowball sampling is an example of non-probability sampling method?

a. True

b. False

True

When designing a study, a very large sample size is the only factor that guarantees

the quality of the results.

a. True

b. False

False

Which factor primarily determines which statistical techniques a researcher can

use?

a. Sample size

b. Metrics of measurement (nominal, ordinal, interval, ratio)

c. Study duration

d. Researcher preference

Correct answer is B

Temperature (measured in degree F or degree C) is an example of which metrics of

measurement?

a. nominal

b. ordinal

c. interval

d. ratio

Correct answer is C

Public health researchers often combine multiple survey questions into a scale or

index to measure complex constructs.

a. True

b. False

True

Which correctly matches reliability and validity?

a. Reliability = applies to scales; Validity = applies to surveys

b. Reliability = easy to calculate; Validity = easy to interpret

c. Reliability = stable results; Validity = measures correctly

d. Reliability = measures correctly; Validity = stable results

Correct answer is C

Public health researchers may never use self-reported or proxy measures because

they are always unreliable.

a. True

b. False

False

Which of the following is an example of qualitative data analysis method?

a. Logistic regression

b. Meta analysis

c. Content analysis

d. Case-control study

Correct answer is C

A qualitative data collection method where you have a small group discussion about

a particular research topic is best known as…

a) Unstructured interview

b) Semi-structured interview

c) Focus groups

d) Participant observation

Correct answer is C

In qualitative research, a strategy that uses multiple investigators, theories, data source, and methods to provide a more comprehensive understanding of a phenomenon, to overcome the intrinsic bias that comes from single-methods/observer/theory, is known as…

a. Triangulation

b. Emic

c. Quantification

d. Content analysis

Correct answer is A

Survey research usually uses questionnaires, and conducts quantitative data

analysis.

a. True

b. False

True

You set up an online survey, so that…

1. You first ask a question about smoking status, and

2. Anyone who answered NO would skip all questions related to smoking (e.g., type

of cigarette used, frequency of smoking, number smoked per day, etc)

This is known as…

a. Skip logic

b. Manual of Operations (MOP)

c. Pilot test

d. Incentives

Correct answer is A

Which of the following is NOT one of the three common survey modalities?

a. Self-administered

b. interviewer-administered

c. Computer-administered

d. Context-administered

Correct answer is D

Which of the following is NOT one of the popular online survey platforms discussed

in class?

a. Survey Monkey

b. Qualtrics

c. Google Forms

d. Polloria

Correct answer is D

This is the affirmative agreement of a minor to participate in research, given after

the study has been explained in age-appropriate or understandable terms. This

shows the child participant’s voluntary willingness to take part in research,

provided that they already have their parent or legal guardian’s consent. This is

known as…

a. Active consent

b. Passive consent

c. Assent

d. Dissent

Correct answer is C

The primary purpose of program evaluation is to provide evidence-based

information for stakeholders.

a. True

b. False

True

When doing program evaluation, outcome evaluation tends to only occur at early

stages of research.

a. True

b. False

False

The type of program evaluation that assesses program context, reach of the

program, dose delivered and received, fidelity to the original program plan,

implementation, and recruitment… essentially asking the question “are things going

as intended/expected?” is known as…

a. Formative evaluation

b. Process evaluation

c. Outcome evaluation

d. Impact/economic evaluation

Correct answer is B

Which of the following is NOT an example of economic evaluation?

a. Affordability analysis

b. Costs of unmet needs analysis

c. Content validity analysis

d. Cost utility analysis

Correct answer is C

Which of the following is a correct description of Community Based Participatory

Research (CBPR)?

a. It is a type of research intended to aid international policy decisions

b. It allows academics to conduct research for communities, where community

members take on the role of passive observers

c. It tends to have worse outcomes compared to traditional non-CBPR research

methods

d. It is particularly useful for gaining entry into a community, empowerment,

and long-term change

Correct answer is D

The core principles of Community-Based Participatory Research (CBPR) can be

summarized as (1) ethical engagement and (2) community empowerment

a. True

b. False

True

True or False? - Public health researchers may

never use self-reported or proxy measures because they are always unreliable

False

True or False? - Public health researchers may never use self-reported or proxy measures because they are always unreliable.

False

True or False? - Public health researchers often combine multiple survey questions into a scale or index to measure complex constructs.

True

• Which of the following primarily determines which statistical techniques a researcher can use?

Sample size

Metrics of measurement (nominal, ordinal, interval, ratio)

Study duration

Researcher preference

Metrics of measurement

Provide explanation and/or examples of each metrics of measurement (nominal, ordinal, interval, ratio)

• Nominal = e.g., gender, ethnicity, color of eyes

• Ordinal = e.g., self-rated health (“very good”, “good”, “poor”)

• Interval = e.g., temperature (zero is just another point)

• Ratio = e.g., body weight (zero means absence)

True or False? – Observational research is the most common type of research reported in public health

True

• Which of the following is NOT an observational study design?

RCT

Longitudinal

Case-control

Multi-year cross-sectional

RCT

What is the biggest difference between cross-sectional and longitudinal research designs?

Number of times data is collected

This study uses a retrospective research design, originally created for epidemiologic studies. This is very useful for studying rare diseases. This does NOT rely on samples acting as their own controls. This is known as…

Case control study

• Pick out the one that is NOT a true experimental design, and explain why (what is actually is).

Posttest control group design

Pretest-posttest control group design (typical RCT)

Interrupted time series design (ITSD)

Repeated Measures Design

Interrupted time series design (ITSD) = quasi-experimental

True or False? – Studies that use representative data generally have high external validity.

True

• Which of the following words best describe what sampling is all about?

Representativeness

Size

Accuracy

Randomness

Representativeness

This is a sampling method also known as the “chain referral” sampling. It is the most useful for hard to reach or hidden populations. You are essentially getting participants to help recruit more participants. This is…

Snowball sampling

Simple random, Systematic random, Stratified random, and Cluster sampling are all examples of…?

Probability sampling

• Name the second criteria considered when we make sampling decision.

Hint: Sampling (1) element, (2) ____, and (3) technique

Sampling frame

True or False? – Survey research usually uses questionnaires, and conducts quantitative data analysis

True

• You set up an online survey, so that…

1. You first ask a question about smoking status, and

2. Anyone who answered NO would skip all questions related to smoking (e.g., type of cigarette used, frequency of smoking, number smoked per day, etc)

This is known as…

Skip Logic

This is something that only legal adults with no mental/physical impairment can give. A child is able to give assent, which is somewhat similar to this. This refers to…

Consent

What is a popular online survey platform that also has some capacity participant recruitment? The company behind this platform, Q_____, LLC, was founded in 2002, and has co-headquarters in Seattle, Washington, and Provo, Utah, in the United States.

Qualtrics

• Name two of the three survey modalities

hint: “____-administered”

Self, interviewer, computer-administered

True or False? – Primary purpose is to provide evidence-based information for stakeholders

True

True or False? – Process evaluation tends to only happen at the end of a program

False

What is the earliest stages of program evaluation, where you try and assess the feasibility of a program?

Formative Evaluation

In the context of program evaluation, affordability analysis is an example of what?

Economic Impact Analysis

Name two of the four types of program evaluation

(1) formative, (2) process, (3) outcome, (4) impact/economic

This is a type of research conducted by and for stakeholders. It allows academics to conduct research with communities, at highest level of involvement. It is particularly useful for gaining entry into a community, empowerment, and long-term change. This is known as…

CBPR (Community-Based Participatory Research)

Nine Steps to Research

1. Identify priority population

2. Formulate research questions

3. Determine if research is observational or experimental

4. Select a parasimonious research design

5. Determine variables measured

6. Select sampling method

7. Implement study protocol

8. Clean and analyze data

9. Answer research question and disseminate findings

Sections of a Research Paper

Abstract

Introduction

Methods

Results

Discussion

References

Epistemology

Distinguishes true knowledge from false knowledge

P.I.C.O

P-Patient/Population

I-Intervention/Treatment

C-Comparison

O-Outcome

Exploratory Research

Research that is mostly descriptive in nature, intended to identify basic health issues that serve as a springboard into further inquiry

Conclusive Research

Determines cause and effect

Types of Observational Research Designs

1. Cross-Sectional

2. Longitudinal

3. Successive Independent

4. Cohort Sequential

5. Case-Control

6. Case-Crossover

Cross-Sectional

Only a single moment in time serves as the measurement window

Longitudinal Design

Study participants are followed for over a defined period of time

Case Control

Roots in epidemiology. Cases have the disease and controls do not.

True Experimental Designs

Post-test Control Group

Pretest-Post-test Control Group

Matched Pairs Design

Repeated Measure Design

Crossover Trial Design

Post-Test Control Group

Weakest of the true experimental designs. Between subjects designs

Pre-test-Post-test Design

Includes pre-test measurements. Between subjects designs. Susceptible to low external validity

Matched Pairs

Subjects are matched by characteristics then persons from each group are paired randomly to an assigned group. Between subjects design

Repeated Measures Design

Subjects serve as their own controls. Reduces error variance. Within subjects design

Crossover Trial Design

Reduces carryover effects. Uses counterbalancing (half subjects randomly assigned to receive level 1 then level 2 IV while other half receives vice versa). Within subjects design

Quasi-Experimental Designs

Allows casual inference despite not being truly experimental. May use observational data

Types of Quasi-Experimental Designs

Non-equivalent Control Group Design (NCGD)

NCGD Post-test Only

NCGD Pre-Test Post-Test

Interrupted Time Series Design

Non-equivalent Control Group Design (NCGD)

Includes a control group, but does not involve randomization. Control group is closely matched to treatment group but not enough to be considered “equivalent”

NCGD Post-Test Only

Used when unable to incorporate pre-test measurements, but can create a control or comparison group. Has many threats to internal validity

NCGD Pre-test Post-Test

Implements a pre-test measure to assess differences between groups and changes within groups. Can use longitudinal or multi-wave cross sectional data

Interrupted Time Series Design

Strongest of the quasi-experimental designs. Evaluates long-term effects of an intervention. Collects a series of data points related to the dependent variable over time.

Precision

How consistent repeated measurements are with each other

Accuracy

How close a measurement is to the true-accepted value

Reliability

How consistently a tool or test measures something across time, raters, or items (does it give stable results)

Validity

How well a tool actually measures what it’s supposed to measure

Five Characteristics of Qualitative Research

Is it naturalistic

The data are descriptive

Concerned with process

Utilizes an inductive approach

Goal is to find meaning

Four Purposes of Qualitative Research

Exploratory Research

Enhance the depth and validity of quantitative inquiries

A precursor to implementing health promotion programs

An alternative method when other methods are

preclude