BBH 310 Exam 2 questions
When doing a within-subjects comparison without a control group, what is one thing that is not likely to be a confounder? Natural recovery over time Expectations related to the study "Person variables"
"Person variables"
Researchers test the common belief that Vitamin C speeds up recovery from the common cold. In this study, a group of subjects who all have the cold are randomly assigned to either take a Vitamin C pill or a placebo. The time to recover each time is measured. This design is a ______-subjects study Within Between
Between
When we examine the link between a "person variable" and a health outcome, we would use an _______ design. Experimental Observational
Observational
Students in BBH 411W are running a little study. When participants get to their station, they ask them what month they were born in. If the month was Jan-June, they go in the control group. If the month was July-Dec, they go in the treatment group. This design is Experimental Quasi-experimental
Quasi-experimental
Three of these are key rules for deciding if a single study is providing evidence for causation, and one is not. Which is the one that is not a key rule for deciding if a study can be interpreted as causal? Researchers use statistical control to account for effects of confounders Researchers assign subjects to different levels of the independent variable at random Researchers manipulate the independent variable Researchers use experimental control to hold confounders constant
Researchers use statistical control to account for effects of confounders
What is one advantage of a cross-over experimental design (see notation below) in comparison to a pre-post study? Group 1: R O X O O Group 2: R O O X O You have data to observe what happened at baseline and thus see within-subjects differences You can manipulate the intervention of interest You have data to estimate outcomes in the absence of an intervention despite a similar amount of time
You have data to estimate outcomes in the absence of an intervention despite a similar amount of time
which design would one most likely use to study the impact of interventions to increase vaccination initiation? factorial crossover pre-post between subjects quasi
between subjects quas
observational studies are "better" than experimental studies in the sense that they: can test the effect of a new treatment on a certain health condition can investigate risk factors of diseases of interest more in-depth overall they require less effort to complete than experimental studies do
can investigate risk factors of diseases of interest more in-depth
a researcher team begins following 1,000 students from Penn state to track their self-reported alcohol consumption as well as the physical activity levels to see if there is any association between these two variables. this would be an example of a(n) Case study ecological study ecological intervention case series
case series
a study follows 6 individuals with an extremely rare immondeficiency disease.... case study EMA ecological intervention case series
case series
For which study design would it NOT make sense to use a random sample? case study cross-sectional study case-control study cohort study experimental study
case study
a researcher studies whether a therapy works for reducing BP. aging is the single largest risk factor for high BP, so its important to make sure that age is not a confounder in the study. what is the most appropriate way for the researcher to handle this issue? -control for age experimentally, using random assignments to balance out age effects -control for age statistically, using statistical analysis to balance out age effects
control for age experimentally, using random assignments to balance out age effects
case-control studies are particularly poor for --, but are particularly strong at deciphering potential causes for---. -detecting person-variables; common diseases -detecting subjective exerpeicne; common diseases -detecting person variables; rare diseases -detecting subjective experiences, rare diseases
detecting subjective experiences, rare diseases
compared to other types of studies, what is a particular advantage of a cross- sectional study? -ability to test all kinds of variables -no having to worry about random assing - low false neg rates -ease of selection of random sample
ease of selection of random sample
in maryland, all restaurants throughout the entire state went smoke free. this would be an example of a case stuyd ecological study ecological intervention case series
ecological intervention
imagine a randomized trial is done to see if exercise improves mood. The control group gets no exercise classes, and the treatment ground gets exercise classes. Mood outcome is asked face to face at the end. an increase in odd attributes to expectations abut the exercise experience (not necessarily the physical activity itself) would be an: - not a confounder -masking confounder that makes exercise seem less beneficial -exaggerating confounder that makes exercise seem more beneficial
exaggerating confounder that makes exercise seem more beneficial
a researcher abbreviates the description of the research design like this: R O X O R O O Is this design experimental or observational? experimental observational both
experimental
the image shown above is called a --- and it is often included in a --- forest plot, meta analysis forest plot. systematic review funnel plot, narrative review funnel plot, primary research article
forest plot, meta analysis
Most laboratory experiments are designed to have good ____ validity and have less concern for ____ validity, while clinical trials are willing to sacrifice a little bit of ____ validity to improve the ____ validity. internal, external, external, internal external, internal, internal, external external, internal, external, internal internal, external, internal, external
internal, external, internal, external
one major strength of a narrative review is that: it allows the author to bring together the info in a way that best serves the point the author is trying to make there is little to no publication bias or reviewer bias it rids any problems with confounders in the primary research studies they work best when there is a specific topic areas
it allows the author to bring together the info in a way that best serves the point the author is trying to make
this type of secondary research compiles the quantitative data from many studies that used similar methodologies and conducts a large analysis on a specific topic: primary research article systematic review narrative review meta analysis
meta-analysis
A researcher abbreviates the description of the research design like this N O O O Is this study observational or experimental? -Experimental observational
observational
which of the following is not a strength of conducting a simple cohort study? -studying risk factors for common diseases --potential for confounders -able to see precedence -can examine variables that would be unethical to manipulate
potential for confounders
A retrospective case-control study is published reporting a statistically significant relationship between anger at one's spouse and risk of automobile accidents. However, a critical reader doubts whether the anger observed in the study actually caused the automobile accidents, because it's not established whether the anger actually occurred before the accidents or after the accidents. Thus, the research findings are called into question over the issue of: Plausibililty Precedence Plausibililty Association
precedence
a retrospective case control study is published reporting a statistically significant relationship between anger at ones spouse and risk of automobile accidents. however, a critical reader doubts wether the anger observed in the study actually caused the automobile accidents, because its not established whether the anger actually occurred before the accidents or after. thus, the research findings are called into question over the issue of: plausibility precedence parsimony association
precedence
if you're uncertain of the evidence quality in a particular research area and want to read criticisms of the study design, measurements, sample choice etc for a specific topic of health research the best type of research manuscript to read is: primary research article secondary tertiary
secondary
the weakness of a case study would be that: they are usually extremely detailed sometimes they can attempt to infer cause and effect rel from single events they usually only provide insight on common diseases or situations
sometimes they can attempt to infer cause and effect rel from single events
Which variable is most likely to be manipulated during an experiment? Aggression Stress Age Ethnicity
stress
which of the following is a major strength of EMAs? you can only measure things that are quick and easy to meausre studying realistic situations leads to higher external validity active data collection can never lead to bias you can use traditional statistics to analyze EMA data
studying realistic situations leads to higher external validity
a paper is published in the journal "diabetes care". The authors describe in detail their strategy to find results from all the papers published between.... meta analysis systematic review narrative review primary research article
systematic review
if you're interested in reading summaries of the take home messages about the state of science as a whole that are specifically geared toward lay persons, administrators, technical workers etc, who don't need to get into al the details of a study, the best research manuscript is: primary secondary tertiary
tertiary
which of the following is a major weakness of pre-post studies: they may contain maturation effects they may contain random error they do not provide precedence they do not contain any intervention
they may contain maturation effects
A researcher is interested in examining the negative effect of chocolate consumption on cardiovascular risk, based on the high sugar and fat content of most chocolates. She decides that Easter, a time of increased chocolate consumption, is the perfect time to test this. She tests cardiovascular risk markers in her participants both 2 days before and 2 days after Easter, and asks for reports about how much chocolate-based easter candy participants ate. Contrary to her hypothesis, she does not find a greater decrease in cardiovascular health among individuals who reported eating more chocolate. A colleague of her points out that chocolate contains flavonoids, which might benefit cardiovascular health. In this researcher's study, flavonoids represent A confound Not a confound, because flavonoids are not associated with the predictor Not a confound, because flavonoids are not associated with the outcome Not a confound, because flavonoids were not extraneous to the original hypothesis Not a confound, because there was no significant association
A confound
A popcorn salesman/amateur researcher studies the effects of watching funny movies on fatigue by surveying people on their energy level as they walk out of the movie theater on a scale of 1-10, depending on whether the movie was a comedy or a different type. He thinks a good laugh will reinvigorate the psyche and dispel fatigue. However, he is concerned about time of day effects--while he was collecting data, the comedies tended to more often be earlier in the day than the non-comedies. Also, as expected, fatigue does change with time of day. Thus, time of day is -A confounder -Not a confounder because it was a mechanism of the hypothesis of interest - Not a confounder because it was unrelated to the predictor of interest -Not a confounder because it was unrelated to the outcome of interest -Not a confounder for multiple reasons above
A confounder
when using the term "ecological" within a study, it means that: measurements are taken in real life/the real world measurements are taken in a persons regular daily environment it is relative to nature Both A &B are correct
Both A &B are correct
What type of study is most suitable for studying risk factors for a very rare disease? Cross-sectional Cohort Case-control Clinical trial Lab experiment
Case-control
What design will produce the most trustworthy evidence if you are studying a risk factor for a common disease, where this risk factor is harmful and cannot be manipulated? Case-control study Cross-sectional study Cohort study Case study Experimental study
Cohort study
A researcher studies whether a therapy works for reducing blood pressure. Aging is the single largest risk factor for high blood pressure, so it's important to make sure that age is not a confounder in the study. What is the most appropriate way for the researcher to handle this issue? Control for age experimentally, using random assignment to balance out age effects Control for age statistically, using statistical analysis to balance out age effects
Control for age experimentally, using random assignment to balance out age effects
A group of BBH 411W students put together a set of questions and we email it out to students in BBH 310 (this happened recently). What is the research design for this project? Cohort Case-control Cross-sectional Lab experiment Clinical trial
Cross-sectional
A research group collects saliva from people who are watching popular action movies during their first showing. The researchers do not decide who shows up to the movies or what is the content of the movies, they just recruit whoever shows up at whichever movies come out during their study time frame. They test the hypothesis that testosterone will be elevated at the end of movies where the hero "wins"/survives and testosterone will be lower at the end of movies where the hero dies a tragic death or fails on a major part of his/her mission. This design is a Case study Cross-sectional Cohort Lab experiment Clinical trial
Cross-sectional
Gallup is an organization that does polls (brief survey studies) of different populations to ask them questions about important issues of our day. Their goal is to figure out what people think. In 2014, they found that many Americans were worried about getting Ebola. This study is a Case study Cross-sectional Cohort Clinical trial Lab experiment
Cross-sectional
Adding a baseline measurement of the outcome variable to a clinical trial helps us to Prevent sampling bias in convenience sampling Determine if that variable changed after the treatment Eliminate the need for a randomized control group
Determine if that variable changed after the treatment
Adding a baseline measurement of the outcome variable to a clinical trial helps us to Eliminate the need for a randomized control group Determine if that variable changed after the treatment Statistically control for confounding third variables You Answered All of these
Determine if that variable changed after the treatment
Researchers compare an exercise intervention to an education-about-exercise-but-no-actual-exercise control group. They measure outcomes at the end of each session of exercise or each classroom activity depending on which arm of the trial it's in. The exercise is done in a gym, while the education group goes to a classroom. Since the researchers are doing random assignment to the exercise intervention and using the control group, they plan to interpret the differences in the outcomes as being caused by the intervention. However, they should watch out for which confounders? Differences in expected benefits and also differences that may relate to the different settings between the groups Differences in ethnicity and general preference for exercise between the groups Differences in family history of disease and early-life vaccination between the groups
Differences in expected benefits and also differences that may relate to the different settings between the groups
Some schools decide to change their policies on recess to increase outdoor play by 30 minutes and other schools keep recess time the same. A researcher evaluates changes in body mass index in both groups of schools to determine whether there are health improvements in students with increased recess time. What confounder is a particular concern for this quasi-experiment that would not be a problem if it were a true experiment? Different operational definitions of BMI in different schools Different student characteristics in schools that change their policies Random variation may cause type 1 or 2 error
Different student characteristics in schools that change their policies
Case-control studies differ from cross-sectional studies because they Don't have a randomized control group Can't have more than one time point Don't have the same sampling procedures for all participants Can collect information about many aspects of a person's life
Don't have the same sampling procedures for all participants
Researchers investigate the correlation between height (tallness) and blood estrogen levels. They analyze their data two different ways, first including all people in their sample (both genders) and second including only women. The first results show a significant correlation between height and estrogen levels (taller people have lower estrogen). The second results show no significant correlation between height and estrogen levels. They report that gender is a confounder that is _____ the association between height and estrogen levels. Exaggerating Masking
Exaggerating
Randomized clinical trials testing specific psychotherapy and behavioral interventions are almost always confounded by Expectation effects Hypothesized effects Socioeconomic status Unexpected health outcomes
Expectation effects
Rats are assigned at random to live in either a normal cage or in a huge cage filled with lots of other rats and setups for activities. Next, their preference for cocaine is measured. Case study Cross-sectional study Experimental study Cohort study Case-control study
Experimental study
Which of the following is generally true? Experiments are better for eliminating ethical conundrums than observational studies Experiments are better for eliminating demographic confounders than observational studies Experiments make it easier to do random sampling of the population than observational studies Experiments are better for studying the health consequences of violence than observational studies
Experiments are better for eliminating demographic confounders than observational studies
Cohort studies are a much better choice than case-control studies when the predictor is __________ and the outcome is relatively ____________. (Choose the option where the advantages of cohort studies over case-control are most extreme.) Hard to remember correctly, rare. Easy to remember correctly, rare. Hard to remember correctly, common. Easy to remember correctly, common.
Hard to remember correctly, common.
A researcher studies the relationship between consumption of fructose and diabetes. The researcher has three variables about each participant: how much fructose the person usually eats (estimated from surveys about the person's typical diet), the person's age, and diabetes status (whether the person says they have been diagnosed with diabetes or not). The researcher hypothesizes that consumption of fructose causes a fatty liver which screws with some hormones that then leads to insulin resistance which then progresses to full blown diabetes. The researcher finds that age is associated with risk of diabetes (older people are more likely to have diabetes) but not with fructose intake (people of all ages have similar fructose intake). Based on these results, age is likely -Not a confounder because it is not related to the predictor -A confounder - Not a confounder because it is not extraneous - Not a confounder because it is not related to the outcome -all of these answers apply
Not a confounder because it is not related to the predictor
Imagine that researchers conduct a cohort study, following children from early life to adulthood. Researchers recruit one randomized sample of kids from a school district, ask them many questions about their screen time, and then monitor whether or not they develop obesity over the next few years. The researchers find that there is some attrition from the group - each year, some students move away and live in different places. They find that attrition is more common among children with more screen time, compared to among children with less screen time. They find that attrition is not correlated with body weight. In this study, attrition is Not a confounder of the association between screen time and body weight because it is not related to the outcome Not a confounder of the association between screen time and body weight because it is not related to the predictor Not a confounder of the association between screen time and body weight because it is not extraneous A confounder of the association between screen time and body weight
Not a confounder of the association between screen time and body weight because it is not related to the outcome
Three of these are key rules for deciding if a single study is providing evidence for causation, and one is not. Which is the one that is not a key rule for deciding if a study can be interpreted as causal? Researchers assign subjects to different levels of the independent variable at random Researchers use experimental control to hold confounders constant Researchers manipulate the independent variable Researchers use statistical control to account for effects of confounders
Researchers use statistical control to account for effects of confounders
Which is true about extraneous variables? -They only trick the researchers into misinterpreting the results when they relate to both the independent variable and the dependent variable. - They can trick the researchers into misinterpreting the results when they relate to either the independent variable or the dependent variable. -They always trick the researchers to make insignificant things look significant and vice versa.
They only trick the researchers into misinterpreting the results when they relate to both the independent variable and the dependent variable.
When is it more appropriate to do an experimental study than an observational study: -To study the health benefits of a type of psychotherapy -To study the health effects of cocaine addiction - To study the health differences between genders - None of these are appropriate for experimental designs
To study the health benefits of a type of psychotherapy
Imagine that we have two studies investigating the effects of a therapy on an outcome, one that allows people to volunteer for the therapy (so it's really more of an observational study) and one that is truly experimental (a randomized clinical trial). What type of confounder is consistently more of a problem in the observational design than the randomized clinical trials? Unequal distribution of demographic variables between the individuals who receive the treatment and the individuals who don't Unexpected health outcomes influencing a sick population Different treatment and measurement times for the individuals who receive the treatment and the individuals who don't
Unequal distribution of demographic variables between the individuals who receive the treatment and the individuals who don't
Listed below are three true reasons why observational studies have advantages over experimental studies. However, one statement is a true reason why experiments are better than observational studies--which is it? You have fewer problems with confounding variables You have fewer problems with the ethics of manipulating variables You can study the effects of dangerous drugs of abuse You can study the effects of dangerous drugs of abuse
You have fewer problems with confounding variables
Listed below are three true reasons why observational studies have advantages over experimental studies. However, one statement is a true reason why experiments are better than observational studies--which is it? You can study the effects of dangerous drugs of abuse You have fewer problems with confounding variables You can study the effects of ethnicity You have fewer problems with the ethics of manipulating variables
You have fewer problems with confounding variables
A researcher interested in studying stress asks participants to record their feelings of daily stress through a survey question that is texted to them every night for one week. This would be an example of --data collection? active passive both neither
active
attrition is a confounder when -it threatens internal validity -it occurs in people with different levels of health, and in a way that relates to the predictor variable -it affects the overall conclusions of the study -only 2 of these -all of the above
all of the above
common tertiary sources can include: textbooks databases manuals both a and c all of the above
all of the above
in order to conclude that x causes y in an experimental study, what must you do to ensure that there is an actual casual relationship? randomly assign groups manipulate the independent variable control for extraneous factors two of these all of the above
all of the above
which of the following is a strength of a meta-analysis? it yields quantitative results from several studies in clear results it don't correctly, it can provide the most accurate estimate of the benefit to be gained from a therapy of insight to risk factors there is reduced reviewer bias all of the above
all of the above
A researcher studies how choices parents make about sending kids to preschool and grade school can influence infectious illness. In one project, she tests the hypothesis that children who are homeschooled for pre-K and grade school are at greater risk for getting infectious mononucleosis ("mono") in college. She recruits senior college students, asks about history of schooling when young, and asks whether they got mono in college. This research design is: - An observational study -Involves both observational and experimental components -Neither -An experimental study
an observational study
which would be least likely to serve as the IV in a crossover experimental design? invoking a stress response providing a druge to treat cancer behavioral therapy to improve sleep providing a drug to help improve cog
behavioral therapy to improve sleep
what is this an example of?" in this paper, we describe a series of cases that exemplify a temporal associating btwn exposure to violence and the precipitation of asthma....in the first three cases..."
case study
a fascinating study observed a 24 year old patient who had never experienced any pain in her life. this would be an example of a : case stuyd ecological study ecological intervention case series
case stuyd
dentify the study design. Lee et al. (2015) write: "We aimed to determine whether changes in metabolic health status or obesity over time have any effect on the risk of future diabetes. Nondiabetic individuals (n = 2692) from a study with baseline and 2 follow-up examinations at 4-year intervals were included. Individuals were classified as "obese" or "nonobese" using a body mass index of 25 kg/m2 as a cut-off. The risk of diabetes at year 8 was assessed according to changes of metabolic health status between year 0 and 4. Changes in metabolic health status were an independent risk factor for future diabetes in nonobese individuals, whereas general obesity had a greater contribution to the risk of obese individuals developing diabetes." case study cross-sectional study case-control study cohort study experimental study
cohort study
a research lab is interested in studying the assoc between freuquency of physical activity and overall current CD risk in a representative sample of American citizens. this suggests that they will want to use which study design? case study cross sectional case control cohort experimental study
cross sectional
group A patients were administered a drug for 8 weeks followed by a wash out period, and the another drug for 8 weeks. this order of drug administration was reversed in group B factorial crossover pre-post exp between-subj quasi design
crossover
an article that does NOT describe how they found their info, but summarizes empirical research findings on a particular subject in a flexible way would be considered a primary research article a systematic review narrative review meta-analysis
narrative review
originally, observational research on hormone replacement therapy for women showed benefits for reduction of heart disease risk. However, later research using randomized clinical trials to eliminate several demographic cofounders showed no benefit. The randomized clinical trials had better-- than the observational cohort studies association plausibility parsimony precedence
parsimony
an article that describes the methods and results of one experimental study in depth could best be described as a : primary research article secondary tertiary
primary research article
why are person variables less likely to confound experiments as opposed to observational studies? -the potential confounders are measured and stat controlled for in experiments, but not in the observational studies -manipulation of the IV forces everyone to be equal in experiment, but not in observational studies -random assign balances the confounding variables between the groups in experiments -the research team experimentally controls for these variables by assigning them to be equal across the experiment
random assign balances the confounding variables between the groups in experiments
when an author intentionally or unintentionally only includes the studies that support the argument they're tying to make, while neglecting available studies that contradict the argument, this would result in what kind of bias? publication bias reviwer bias sampling bias recall bias
reviwer bias
which of the following is NOT a strength of ecological interventions? the study of the exosystem the study of multiple contexts that influence health behaviors the assignment is always random they include some kind of control group
the assignment is always random
what are some of the benefits to doing a time series analysis? they can predict short-term trends based on previous patterns they don't take very long to complete there are very few data point collections the data can usually be applicable to the general public
they can predict short-term trends based on previous patterns
one major weakness of a systematic review is that: it is less representative of the truth than a single empirical study it does not include set guidelines for the review they can suffer from publication bias they work best when there is a specific research question
they can suffer from publication bias
a particular strength of a quasi exp are that researchers can examine the effect of unethical behavior true false
true
in general, it is good to be skeptical of sources that have financial incentives or monetary influences true false
true
one benefit of ecological studies is that they can test to see if particular variables make a difference in explaining why some places are different from others in the real world true false
true
specifically, what is a major difference between a true experiment and quasi experiment? true experiements do not draw from a random sample, whereas quasi do true experiments have fewer confounders than quasi-experiments researches can detect precedence in true-experiments but not quasi experiments there are no difference
true experiments have fewer confounders than quasi-experiments
which types of error are eliminated or drastically reduced by aggregating different study results together? masking and exaggerating confounders type one and type two statistical errors measurement and sampling bias all of the above
type one and type two statistical errors
A nutritionist is studying childhood dietary factors that might influence risk of asthma in Mexico. The researcher finds a statistically significant relationship between asthma and consumption of a food called poc chuc. This food is a local tradition in the Caribbean part of Mexico where most people are descended from Mayan people. Could some person variables confound the relationship between the food and asthma? yes no
yes
a team of community health promotion researchers implement an intervention in a series of small towns, and they have another series of small towns that don't receive the intervention. The intervention involves a bunch of new free after school programs for kids that are advertised to everyone in the intervention communities- the researchers hypothesize that the time spent in these programs will bring people together and keep kids away from risky behaviors. The control receives no programs or advertisements. Outcomes are assessed by surveying people in each of the towns (both intervention and control) about their feelings about the community and their health behaviors. peoples self-reported outcomes can generally be influenced by placebo effects. the researchers are wondering if some kind of expectation effects might confound the association in their design. -no, expectation is not a confounder because it is not extraneous to the original hypothesis -yes, it looks like a confounder -no, expectation is not a confounder bc it is not related to the Dv -no, expectation is not a confounder bc it is not related to the IV
yes, it looks like a confounder
which of the following is NOT a weakness of a case-control study? you cant study rare cases there may be different selection biases between cases and controls you can have confounding due to lack of random assignment
you can't study rare cases
