BDS (There were only 24 Questions in Practice Exam)

An educational scene depicting Bayesian statistics and behavioral data science, featuring graphs, charts, and diverse individuals collaborating over data analysis in a modern classroom setting.

Bayesian Behavioral Data Science Quiz

Test your knowledge in Bayesian reasoning and behavioral data science with our comprehensive quiz featuring 24 thought-provoking questions. Designed for students and professionals alike, this quiz explores key concepts, methodologies, and principles that form the foundation of Bayesian statistics and behavioral contact networks.

24 multiple choice questions
Covering critical topics like diagnostic tests, symptom networks, and statistical methodologies
Perfect for researchers, students, or anyone interested in advanced statistical concepts

24 Questions6 MinutesCreated by AnalyzingData45

In Bayesian reasoning, if the “predictive updating factor” for hypothesis A versus hypothesis B equals one, then

Hypothesis A predicts the data better.

Hypotheses A and B predict the data equally well.

Hypothesis B predicts the data better.

In a behavioral contact network, nodes represent _______ and links represent _______ .

People; contacts.

Symptoms; causal relations.

People’s locations; distances.

The phenomenon of critical slowing down implies that, before a transition, the state of a system becomes ______ from its previous state, as the system approaches the tipping point.

More predictable

Less predictable

Locally predictable

	Disease	No Disease
+ Test	50	10
- Test	40	90

The table represents the findings of a research project to test the quality of a diagnostic instrument. The numbers in the table represent the number of individuals in the relevant cell; e.g., 50 individuals have the disease and are tested positively. Given this table, the sensitivity of the test in question for diagnosing the disease would be estimated at:

50/90

90/100

140/200

Cognitive test scores are typically positively correlated (the positive manifold). Some researchers explain this through the hypothesis that all test scores are influenced by general intelligence, which causes the scores to become correlated. Here, the positive manifold functions as ____ and the idea that they are all influence by g is ____:

Data; a theory

A phenomenon; a theory

Data; a phenomenon

In the debate of frequentist versus Bayesian statistics, there is no disagreement on whether

Subjectivity should be incorporated in statistical analyses.

Researchers should aim to estimate the probability that a hypothesis is true.

P-values are often misinterpreted in scientific research.

In symptom networks, correlations between symptoms arise because

Symptoms directly influence each other.

Symptom connections influence each other.

Symptoms become more strongly connected over time.

N Lecture 3 on Big Behavioural Data, it was noted that studies of physical distancing suggest that the use of facemasks

Improve social distancing.

Has no effect on social distancing.

Impair social distancing.

Frequentist methodology is organized around an attempt to

Control long-term error rates.

Quantify evidence for the null hypothesis.

Optimize predictive success.

In behavioral data science, data are best thought of as

Representations of observations

Phenomena to be explained

Elements of scientific theory

In the discussion between Bayesians and frequentists, Bayesians typically argue that their approach represents an optimal way of

Rationally learning from data.

Increasing Bayes Factors.

Predicting unseen (“out of sample”) data.

In symptom networks, the concept of hysteresis designates the tendency of symptom networks to

Send out Early Warning Signals before transitions.

Get stuck in a state where symptoms keep activating each other.

Connect different disorders through ‘bridge symptoms’.

The first step in Theory Construction Methodology is to

Develop a plausible theory.

Construct a formal model.

Identify a robust phenomenon.

The speed at which simulation programs of human behavior improve, suggests that in the future we will

No longer need experimental data to study social distancing.

Still need experimental data to inform simulation models for social distancing.

Be able to empirically calibrate simulation models for social distancing against each other.

A Bayesian motivation for the idea that, other things being equal, simpler models should be preferred is the fact that simpler models

Are easier to estimate.

Are more consistent.

Make more precise predictions.

The ROC curve is used to evaluate

The quality of diagnostic tests.

The fit of a diffusion model

The robustness of formal theoretical models

The curvature of rocks

In the behavioral contact networks method, which Tanis and Blanken discussed in Lecture 3 on Big Behavioural Data, the effect of interventions can be assessed by comparing the average _______ across conditions.

Distance between individuals

Contact duration

Number of contacts

If a researcher updates their posterior belief in the hypothesis that a person has a condition, based on a positive test result and the base rate, that researcher is reasoning along the lines of

Frequent hypothesis testing

Bayesian statistics

Likelihood paradigms

In frequentist statistics, p-values can justify the interpretation that

The null hypothesis is true.

There is strong evidence for the null hypothesis.

Given H0, the observed data are improbable.

Estimation approaches to determine symptom networks are based on the statistical technique of

Regression

Cluster analysis

Factor modelling

Bayes’ rule shows that if a hypothesis is very unlikely a priori, then the evidence needed to persuade us of the truth of such a hypothesis has to be very strong. This principle is often condensed in the saying

"extraordinary claims require extraordinary evidence"

"we should not multiply reality beyond necessity"

"simple models should be preferred over complicated ones"

"alles hat ein ende, nur die wurst hat zwei"

A Pizza Plot, as discussed in Lecture 5 on Bayesian inference, is used to evaluate

The predictive success for the null hypothesis versus the alternative hypothesis.

The probability that the null hypothesis is true.

How likely the null hypothesis is relative to the alternative hypothesis.

The ratio of cheese to sauce on an individual slice

BDS (There were only 24 Questions in Practice Exam)

Bayesian Behavioral Data Science Quiz

More Quizzes