BIOSTAT 521: Course Introduction & Study Designs

Flashcards covering course logistics, grading policies, generative AI rules, and foundational concepts of study design and sampling methods from Chapter 1 and Chapter 2.

What are the prerequisites for BIOSTAT 521?

Familiarity with college algebra and calculus.

How is the final course grade weighted in BIOSTAT 521?

Exams count for 75% (25% each across three exams), Homework counts for 10%, and Labs count for 15%.

What is the penalty for late homework submissions in BIOSTAT 521?

Late homework assignments lose 25% credit per day.

What are the four evaluation metrics in the computer lab grading rubric, and what is the starting default score?

The four metrics are Appropriate Use of Statistical Methods, Visual Presentation of Data and Results, Creativity of Analysis (aka 'Taking your analysis further'), and Written Communication. Each lab report starts with a default score of 10/12.

What major survey data source is used across the computer lab modules in BIOSTAT 521?

The National Health and Nutrition Examination Survey (NHANES).

What are the three main rules regarding Generative AI usage for assignments in BIOSTAT 521?

1. Write your own solutions in your own words. 2. Write your own code and understand every line. 3. Disclose your use of generative AI with a brief statement at the bottom of your submission.

What is the primary distinction between an experimental study and an observational study?

In an experimental study, the investigator controls and randomizes who receives the exposure. In an observational study, the investigator does not control the exposure and observes naturally occurring realization of exposures and outcomes.

What is simple random sampling?

A sampling scheme where every individual in the population has an equal probability of being selected for study inclusion.

How does sampling bias occur?

Sampling bias occurs when individuals with certain characteristics are inadvertently sampled with higher probability, producing a sample that is not representative of the intended target population.

Why were the genotype-BMI association estimates from the Genes for Good study considered valid despite non-random recruitment?

Because the demographic non-random recruitment via social media did not bias the biological relationship between genotypes and BMI, yielding effect estimates consistent with larger studies like Locke et al.

What type of sampling design does the National Health and Nutrition Examination Survey (NHANES) utilize?

A complex, multistage, probability sampling design that oversamples specific subgroups (such as African Americans and individuals >60 years of age) and assigns sampling weights.

What multi-step process is illustrated in this sampling diagram?

A four-stage multistage sampling design: Stage 1 samples counties, Stage 2 samples segments, Stage 3 samples households, and Stage 4 samples persons.

What is the defining feature of a cohort study?

A cohort study identifies a group of individuals who do not have the outcome at the start of follow-up and tracks them over time to observe disease onset, establishing temporal ordering.

What study design timeline is depicted in this diagram?

Prospective Cohort Design

What study design timeline is depicted in this diagram?

Retrospective Cohort Design

What was the initial cohort size and inclusion criteria for the Framingham Heart Study when it began in 1948?

An original cohort of 5,209 men and women aged 30 to 62 from Framingham, Massachusetts, who had not yet developed overt symptoms of cardiovascular disease, heart attack, or stroke.

What is a cross-sectional study?

An observational study design in which data are collected from different individuals at a single point in time.

How are participants selected in a case-control study design?

Study participants are recruited based on their outcome status (cases with disease vs. controls without disease), with cases typically over-sampled to increase statistical power.

Which study design is illustrated in this diagram where subjects are sampled based on disease status?

Case-Control Study

Why is a case-control design superior to a simple random sample or cohort design for studying cystic fibrosis?

Cystic fibrosis is a rare disease, making simple random sampling or prospective cohort follow-up inefficient or impractical for observing a sufficient number of cases.

Which study design is best suited for estimating disease prevalence in a population?

A cross-sectional study or simple random sample.

According to the hierarchy of evidence pyramid shown below, which category of studies provides the highest evidence for causation?

Systematic Reviews and Meta-analyses

What was the cohort definition and study design of the Chronic Wasting Disease (CWD) Sportsmen's Feast study?

It was a prospective cohort study where inclusion was defined by having consumed CWD-positive venison at the Sportsmen's Feast in Oneida County, NY on March 13, 2005.