Exam 1 Flashcards
(22 cards)
What is the null hypothesis (H₀) in a t-test or ANOVA?
There is no difference between group means; any observed difference is due to chance.
What statistical test would you use for one categorical independent variable (3+ groups) and one continuous dependent variable?
One-way ANOVA.
What is the purpose of random assignment in experimental design?
To reduce bias and ensure independence of experimental units.
What does a p-value represent in hypothesis testing?
The probability of obtaining a test statistic at least as extreme as the observed one, assuming H₀ is true.
What are two types of error in hypothesis testing?
Type I: Rejecting a true null; Type II: Failing to reject a false null.
Define a ‘closed population’ in the context of mark-recapture studies.
A population in which no births, deaths, immigration, or emigration occur during the study.
What are the assumptions of the Lincoln-Peterson estimator?
1) Closed population, 2) No mark loss, 3) Equal capture probability, 4) Marks don’t affect recapture probability.
What is the Lincoln-Peterson formula for estimating population size (N̂)?
N̂ = (M × C) / R, where M = marked in 1st sample, C = caught in 2nd, R = recaptures.
How do you estimate capture probability (p̂) using Lincoln-Peterson?
p̂ = R / C (R = recaptures, C = total caught in second sample).
What is ‘effective strip width’ in distance sampling?
The width within which all animals would be detected if detection probability were 1.
Name one key assumption of distance sampling.
Animals on the line are always detected with certainty.
In linear regression, what does the slope (β₁) represent?
The change in y (response) for each unit increase in x (predictor).
What does R² represent in regression?
Proportion of variation in the response variable explained by the predictor variable.
What are the assumptions of linear regression?
Linearity, normal residuals, constant variance (homoscedasticity), and independence.
What does AIC stand for, and what is its purpose?
Akaike Information Criterion; selects the most parsimonious model balancing fit and complexity.
What does ΔAIC < 2 between models mean?
The models are similarly supported by the data.
What is the principle behind model selection using AIC?
The principle of parsimony – balancing model fit with number of parameters.
What is a parameter in statistics?
A true (but often unknown) value describing a population.
What is an estimator?
A formula or method used to estimate a population parameter based on sample data.
What is a sampling distribution?
The distribution of an estimator over repeated samples from the same population.
What is sampling bias?
Systematic error that favors certain outcomes, leading to inaccurate estimates.
How does increasing sample size affect precision?
It reduces sampling error and increases the precision of estimates.
