BIOL2001 Introduction to Quantitative Biology
The Complete Study & Assessment Guide · S1 2026BIOL2001 Overview
- Semester 1, 2026
- Level 2 undergraduate course
- 6 units, taught in person
- Weekly R computer workshops
- Assessed entirely in class
What the course covers
Introduction to Quantitative Biology at the Australian National University teaches the quantitative toolkit that biology and ecology rely on. It runs in three parts.
- Assessed by Four in-class practical tests worth 20% each and four Canvas quizzes worth 5% each, all held in tutorials or workshops during semester.
- Biggest share Part III, Statistical Models in Ecology, carries half of the grade through two quizzes and two tests in Weeks 9 to 12.
- How to prepare Practise every method in RStudio as it is taught, because the tests are computer based and ask you to write and run R code yourself.
How BIOL2001 is assessed
| Component | Weight | Format |
|---|---|---|
| In-class Canvas Quiz 1 | 5% | Week 3 tutorial: computer-based quiz on Canvas, R programming |
| In-class Practical Test 1 | 20% | Week 4 tutorial: computer-based test implementing R code |
| In-class Canvas Quiz 2 | 5% | Week 6 tutorial: computer-based quiz on probability and statistics |
| In-class Practical Test 2 | 20% | Week 7 tutorial: about 40 minutes, R programming with probability and statistics |
| In-class Canvas Quiz 3 | 5% | Week 9 tutorial: computer-based quiz on linear models |
| In-class Practical Test 3 | 20% | Week 10 tutorial: about 40 minutes, statistical models |
| In-class Canvas Quiz 4 | 5% | Week 11 tutorial: computer-based quiz on logistic models |
| In-class Practical Test 4 | 20% | Week 12 workshop in 2026: about 40 minutes, statistical models |
The weights come from the ANU Class Summary for Semester 1, 2026, which lists In-class Practical Tests worth 80% in total, four tests of 20% each, and In-class Canvas Quizzes worth 20% in total, four quizzes of 5% each, and states that there is no final examination. An earlier course information page on Canvas lists eight quizzes and a test or assignment in Week 13; the Class Summary, the Canvas quiz pages and the lecture slides all give four quizzes and four tests ending in Week 12. Late submission is not permitted, and a missed test can be deferred only with a medical certificate.
Current BIOL2001 dates
| Date | Item | Control |
|---|---|---|
| 10 March 2026 | In-class Canvas Quiz 1 | Week 3 tutorial, R programming |
| 31 March 2026 | In-class Canvas Quiz 2 | Week 6 tutorial, probability and statistics |
| 21 April 2026 | In-class Practical Test 2 | Week 7 tutorial, about 40 minutes |
| 5 May 2026 | In-class Canvas Quiz 3 | Week 9 tutorial, linear models |
| 19 May 2026 | In-class Canvas Quiz 4 | Week 11 tutorial, logistic models |
| 26 May 2026 | In-class Practical Test 4 | held in the Week 12 workshop |
Dates are as published in the Semester 1, 2026 Canvas quiz and test pages and the Part II and Part III lecture slides. Confirm exact deadlines and submission settings in the live LMS.
What BIOL2001 covers
Fifteen chapters follow the three parts of the course: programming in R, probability and statistics, and statistical models for ecological data.
Values, Data Types and Data Structures in R
Operators · assignment · integer, double, logical and character types · NA, Inf, NaN and NULL · vectors, matrices, lists and data framesThe first week of R
Part I opens with how an R session works: you type a command at the prompt, R reads it, evaluates it and prints the result.
- The read-evaluate-print loop and RStudio
- Arithmetic operators including %/% and %%
- Logical operators and TRUE or FALSE results
- Assignment with the arrow operator
- Integer, double, logical and character types
- NA, Inf, NaN and NULL
Conditional Statements and Loops in R
if and else · compound expressions · for and while loops · tracing a loop · break, next, repeat and switchMaking a program decide and repeat
Without flow control, R runs a script from its first line to its last.
- Programs as ordered instructions
- Conditional if and else statements
- Compound expressions in braces
- For loops over a vector
- While loops and their stopping condition
- Tracing a loop pass by pass
Indexing and Subsetting Vectors, Lists and Matrices
Positive, negative and logical indexes · which() · matrix rows and columns · single and double brackets · the dollar operatorPicking the parts you need
Subsetting is how R code reaches into a structure and takes out the elements a task needs.
- Positive indexes keep positions
- Negative indexes drop positions
- Logical indexes and vectorised comparisons
- Positions of TRUE elements with which()
- Row and column indexes for matrices
- Single brackets, double brackets and the dollar sign
Data Frames: Files, Factors, Merges and Aggregation
read.table and write.table · factor levels · adding and deleting columns · rbind, cbind and merge joins · subset() and aggregate()Working with real tables
A data frame holds a field data sheet in R: each column can have its own type, but every value in a column shares it.
- Data frames as heterogeneous matrices
- Arguments of read.table and write.table
- Selecting rows with logical expressions
- Factors, levels and integer codes
- Adding and deleting columns
- Stacking tables with rbind and cbind
Functions, the Apply Family and R Markdown
User-defined functions · default arguments and the ellipsis · apply margins · lapply and sapply · reproducible reportsWriting reusable code
A user-defined function lets one line stand for a task you repeat.
- Anatomy of a user-defined function
- Default arguments and the ellipsis
- Anonymous functions
- Margins of apply(): rows, columns or cells
- List output from lapply, vectors from sapply
- R Markdown and knitting
Probability Rules, Probability Trees and Bayes' Theorem
Mutually exclusive and independent events · addition and multiplication rules · trees and contingency tables · conditional probability · BayesThe rules behind every probability
Part II begins with the frequentist idea of probability, the proportion of times an event would occur in many repetitions, and the two properties that decide which rule to use.
- Frequentist probability between 0 and 1
- Mutually exclusive versus independent events
- General addition rule
- Multiplication rule and conditional probability
- Probability trees
- Contingency tables and marginal probabilities
Binomial, Poisson and Normal Distributions
Discrete and continuous variables · Bernoulli trials · permutations and combinations · Poisson rates · the empirical ruleThree models for three kinds of variable
A probability distribution assigns a probability to every possible outcome of a random variable.
- Discrete and continuous random variables
- Bernoulli trials and the binomial formula
- Mean np and standard deviation of a binomial
- Permutations, combinations and n choose k
- Poisson rates rescaled to the interval
- Normal density and the empirical rule
Hypothesis Testing With Chi-Square and Fisher's Test
Null and alternative hypotheses · p-values and alpha · goodness of fit · tests of association · Type I and Type II errorsIs the pattern more than chance?
Biological data always carry natural variation and sampling error, so a difference in a sample might be chance.
- Null and alternative hypotheses
- P-values and the alpha threshold
- The five-step testing workflow
- The chi-square statistic
- Goodness-of-fit tests
- Tests of association on contingency tables
Standard Error, z-Scores and the Central Limit Theorem
Parameters and statistics · the n minus 1 divisor · z-scores and normal areas · standard error · sampling distributionsFrom samples to populations
Population quantities such as the mean, variance and standard deviation are parameters; the same quantities computed from a sample are statistics that estimate them.
- Parameters versus statistics
- Sample variance with the n minus 1 divisor
- Standardising values with z-scores
- Areas with pnorm and lower.tail
- Standard error of the mean
- Standard deviation versus standard error
Confidence Intervals, t-Tests and Permutation Tests
Confidence intervals · one-sample, paired and two-sample tests · Welch's test · checking normality · permutation testsIntervals and tests for means
A confidence interval adds a range to a point estimate: at 95% confidence, it has that probability of enclosing the true mean.
- Confidence intervals for a mean
- The t statistic as signal over noise
- One-sample t-test
- Paired t-test on differences
- Two-sample and Welch t-tests
- Normality checks: histogram, QQ plot, Shapiro-Wilk
Simple Linear Regression and Model Assumptions
Least squares · residuals · R-squared · normality, homogeneity, independence and fixed x · transformations and causationA line as a model
Part III opens with regression: describing how a response changes with a predictor, or more precisely the conditional distribution of Y given X.
- Regression as a conditional distribution
- Response, covariate, intercept, slope and error
- Least squares and residuals
- Properties of the fitted line
- R-squared and Pearson's r
- Normality, homogeneity, independence and fixed x
Multiple Regression, Interactions and Model Selection
Dummy variables · ANOVA · partial effects · interaction terms · adjusted R-squared, AIC, BIC and evidence ratiosGroups, several covariates and interactions
This chapter extends the linear model in three directions.
- Binary covariates coded 0 and 1
- Dummy variables and the reference category
- ANOVA for a factor as a whole
- Partial effects in multiple regression
- Interaction terms and separate slopes
- Adjusted R-squared
Logistic Regression and the Generalised Linear Model
Probability, odds and log-odds · the logit link · the S-shaped curve · odds ratios · maximum likelihood and devianceBinary responses
Presence or absence, alive or dead: many ecological responses have only two values, and a straight line fitted to them predicts impossible probabilities below 0 and above 1. Logistic regression solves this by modelling the log-odds, which have no limits, as a linear function of the covariates, then transforming back to probability.
- Why linear models fail for binary responses
- Probability, odds and log-odds
- The logit link and back-transformation
- The S-shaped logistic curve
- Odds ratios from coefficients
- Generalised linear models and link functions
Interpreting Multiple Logistic Regression
Odds multipliers · the divide-by-4 rule · slopes at chosen values · centring and interactions · separationThree scales, one coefficient
A logistic coefficient adds to the log-odds and multiplies the odds by the same amount wherever x starts, but its effect on probability depends on where on the S-curve you are.
- Adding to log-odds, multiplying odds
- Uneven changes in probability
- The divide-by-4 rule
- Evaluating probabilities at chosen values
- Inference with two standard errors
- Several predictors held constant
Poisson Regression, Offsets and Overdispersion
Counts and the log link · multiplicative effects · exposure and offsets · equidispersion · quasi-Poisson correctionCounts as responses
Species richness and abundance are counts, and a straight-line model for counts can predict negative values.
- Counts and the log link
- Why fitted means stay positive
- Assumptions: independence and equidispersion
- Exponentiated coefficients as percentage changes
- Exposure and the offset
- Overdispersion ratio
Part I, Programming in R, covers operators, data types and structures, conditional statements and loops, indexing, data frames, user-defined functions, the apply family and R Markdown.
Part II, Probability and Statistics, covers probability rules and Bayes' theorem, the binomial, Poisson and normal distributions, hypothesis testing with chi-square and Fisher's exact test, standard errors and the Central Limit Theorem, confidence intervals, t-tests and permutation tests.
Part III, Statistical Models in Ecology, covers simple and multiple linear regression, interactions and model selection, logistic regression within the generalised linear model, and Poisson regression with offsets and overdispersion.
How it is taught
Each week brings a two-hour lecture, a one-hour tutorial and a two-hour computer workshop, and the tutorial and workshop are compulsory.
Students bring their own laptop with R and RStudio, and the workshops apply each method to biological datasets such as spider webs, flatfish, beach invertebrates and drinking-water wells.
What the assessment rewards
The whole grade comes from eight assessments held in class: four computer-based practical tests worth 20% each in Weeks 4, 7, 10 and 12, and four Canvas quizzes worth 5% each in Weeks 3, 6, 9 and 11. The course states there is no final exam.
Because the tests ask you to implement R code yourself and to interpret statistical output, this guide gives each block of lectures its own chapter, with original worked examples computed in full and practice items to try before reading the answers.
Worked example · free
Predicting from a fitted regression line and checking one residual
- 1Substitute the snout length: 1.2 + 2.2 × 6 = 1.2 + 13.2 = 14.4 N.
- 1The residual is observed minus fitted: 13 − 14.4 = −1.4 N, so this lizard bites 1.4 N below the line.
- 1The slope says each extra centimetre of snout goes with 2.2 N more bite force on average.
- 1R2 = 0.945 means the line explains about 94.5% of the variation in bite force among the five lizards.
Key terms
- Data Frame
- A table in R whose columns may have different types but whose rows within a column share one type.
- Factor
- R's representation of a categorical variable, stored as integers with a label for each level.
- Conditional Probability
- The probability of one event given that another event has occurred.
- Binomial Distribution
- The distribution of the number of successes in a fixed number of independent two-outcome trials.
- Poisson Distribution
- The distribution of the number of independent events in an interval, with mean equal to its variance.
- Null Hypothesis
- The claim of no effect or no difference that a significance test assumes true until the data reject it.
- P-value
- The probability of results at least as extreme as those observed if the null hypothesis were true.
- Standard Error
- The standard deviation of a sample mean across repeated samples, equal to sigma divided by the square root of n.
- Confidence Interval
- A range that, at a stated confidence level, has that probability of enclosing the true population mean.
- Least Squares
- The estimation method that chooses the regression line minimising the sum of squared residuals.
- Interaction Term
- The product of two covariates, which lets the effect of one depend on the value of the other.
- Odds Ratio
- The odds of an event in one group divided by the odds in another group.
- Link Function
- The function in a generalised linear model that connects the expected response to the linear predictor.
- Overdispersion
- Variation in count data greater than the Poisson model allows, where the variance exceeds the mean.
BIOL2001 FAQ
How is Introduction to Quantitative Biology assessed?
The grade comes from eight in-class assessments listed in the ANU Class Summary for Semester 1, 2026. Four computer-based practical tests are worth 20% each and run in Weeks 4, 7, 10 and 12, and four Canvas quizzes are worth 5% each in Weeks 3, 6, 9 and 11. Part III, on statistical models, carries half of the total.
Does the course have a final examination?
No. Both the Canvas course page and the Class Summary state that there is no final examination. The weekly in-class tests and quizzes replace it, so every mark is earned during semester in the compulsory tutorials and, for the fourth test in 2026, the Week 12 workshop. Missing one of these sessions therefore matters more than in a course with a final.
What happens in the in-class practical tests?
The tests are computer based and run on Canvas, opening five minutes after the class starts and lasting about 40 minutes. The Class Summary says tutorials include in-class tests in which you implement R code individually, so you should expect to write and run code on your own laptop as well as interpret statistical results.
Do I need my own laptop and which software is used?
Yes. You need a laptop with WiFi, ANU VPN access and a battery that lasts at least two hours, with R and RStudio installed before the first tutorial and workshop. Both programs are free. Installation problems should be fixed before Week 1, because there is no time to troubleshoot them in class.
What if I miss a test or submit late?
Late submission is not permitted, and work submitted after the due date without an extension receives a mark of 0. The lectures state that a missed test can be deferred only with a medical certificate, and is then taken in office hours in the same or the following week, or replaced by an oral examination. Extensions are requested through the ANU adjustment app.
Can I use AI tools such as ChatGPT?
The Class Summary states that any use of artificial intelligence must be properly referenced, and that failing to cite the use of generative AI is a breach of academic integrity. The first lecture adds that courses set their own rules on AI tools, so check with the convener before relying on one for any assessed work.
Which statistical models does the course teach?
Part III teaches linear regression for continuous responses, including multiple regression, interactions and model selection with adjusted R-squared, AIC and BIC; logistic regression for binary responses within the generalised linear model; and Poisson regression for counts, with offsets for exposure and checks for overdispersion. Interpreting coefficients on each scale is a recurring skill.
Questions students ask in each BIOL2001 chapter
- What is the difference between a data type and a data structure in R?
- A data type describes a single piece of data, such as integer, double, logical or character. A data structure describes how pieces are organised, such as a vector, matrix, list or data frame. A vector of integers has vector as its structure and integer as its type. Chapter 01 · Values, Data Types and Data Structures in R
- When should I use a for loop rather than a while loop?
- Use a for loop when you know what you are iterating over, such as the elements of a vector or the quarters of a year. Use a while loop when you only know the condition for stopping, such as continuing until a value reaches a threshold. Chapter 02 · Conditional Statements and Loops in R
- Can I mix positive and negative numbers in one R index?
- No. R does not allow positive and negative numbers in the same index and stops with an error. Decide whether you are listing the positions to keep, with positive numbers, or the positions to drop, with negative numbers. Chapter 03 · Indexing and Subsetting Vectors, Lists and Matrices
- What does stringsAsFactors = FALSE do when reading a file?
- It keeps text columns as character vectors instead of converting them into factors. Factors suit grouping variables such as sex or habitat, but not free text or names you may edit, so reading text as text and converting chosen columns later keeps control in your hands. Chapter 04 · Data Frames: Files, Factors, Merges and Aggregation
- What does a function return if there is no return() statement?
- It returns the value of the last expression evaluated in its body. A one-line function such as one that divides count by area therefore needs neither braces nor return(), while a longer function can use return() to make the returned value explicit. Chapter 05 · Functions, the Apply Family and R Markdown
- Are mutually exclusive events independent?
- No. If two events cannot happen together, learning that one occurred tells you the other did not, which is the opposite of independence. Their joint probability is zero, so never multiply the probabilities of mutually exclusive events to find it. Chapter 06 · Probability Rules, Probability Trees and Bayes' Theorem
- How do I know whether to use the binomial or the Poisson distribution?
- Use the binomial when there is a fixed number of trials, each with two outcomes and the same probability of success. Use the Poisson when you count independent events in an interval of time, area or distance with no fixed upper limit on the count. Chapter 07 · Binomial, Poisson and Normal Distributions
- What does a p-value actually tell me?
- A p-value measures how surprising the data would be if the null hypothesis held: the chance of a result at least as far from its prediction as the one observed. It does not tell you how likely the null hypothesis is to be true, nor whether chance made the pattern, and it says nothing about how big or important an effect is. Chapter 08 · Hypothesis Testing With Chi-Square and Fisher's Test
- What is the difference between standard deviation and standard error?
- The standard deviation describes how individual observations spread around the mean. The standard error describes the typical distance between a sample mean and the population mean, and equals the standard deviation divided by the square root of the sample size, so it shrinks as samples grow. Chapter 09 · Standard Error, z-Scores and the Central Limit Theorem
- Which t-test should I use for before and after measurements?
- Use a paired t-test, because the same individuals are measured twice. It works on the differences, so each individual acts as its own control and the variation between individuals is removed from the standard error, making real changes easier to detect. Chapter 10 · Confidence Intervals, t-Tests and Permutation Tests
- What are the four assumptions of a linear regression model?
- The errors are normally distributed for each value of x, they have the same variance for every x, the observations are independent, and the covariate is measured without error, or with error small compared with the response. Each assumption has its own diagnostic plot or check. Chapter 11 · Simple Linear Regression and Model Assumptions
- How many dummy variables does a factor with k categories need?
- It needs k minus 1 dummy variables. The category left without a dummy becomes the reference, and each dummy coefficient is the difference in the mean response between its category and that reference. R builds the dummies itself when the covariate is a factor. Chapter 12 · Multiple Regression, Interactions and Model Selection
- Why not fit a linear regression to presence and absence data?
- A straight line fitted to 0s and 1s predicts values below 0 and above 1, which no probability can take, and the normal error assumption makes no sense for a two-valued response. Logistic regression models the log-odds instead, which keeps every fitted probability between 0 and 1. Chapter 13 · Logistic Regression and the Generalised Linear Model
- What is the divide-by-4 rule in logistic regression?
- The slope of the logistic curve is beta times P times one minus P, which is largest at P equal to 0.5, where it equals beta divided by four. Dividing a coefficient by four therefore gives an upper bound on the change in probability per unit of the predictor, accurate near the midpoint. Chapter 14 · Interpreting Multiple Logistic Regression
- Why use Poisson regression instead of linear regression for counts?
- A linear model can predict negative counts and assumes a constant variance, while counts cannot be negative and their spread grows with their mean. Poisson regression models the log of the mean, so fitted values stay positive, and assumes a variance equal to the mean. Chapter 15 · Poisson Regression, Offsets and Overdispersion
How to prepare for the assessments
Keep pace with the weekly rhythm, because each quiz and test follows the teaching it covers by only a week or two and there is no final exam to catch up for. In Part I, type every lecture example into RStudio, then do the voluntary workshop homework without looking at the solutions; fluency with indexing, data frames and functions is what saves time in the computer-based tests.
In Part II, write the rule or test you are using before every calculation, and practise the five-step hypothesis testing workflow until it is automatic, including the sentence that states the decision.
In Part III, which carries half of the grade, always write the fitted model equation with its coefficients and then substitute values such as zero or the means; practise converting logistic coefficients into odds ratios and probabilities, and Poisson coefficients into percentage changes in the expected count.
Before each test, check that your laptop, R, RStudio and the workshop data files all work, and keep your own commented scripts organised by week.
Your AI Statistics tutor for BIOL2001
Stuck on a hard BIOL2001 question? Sia is AskSia’s AI Statistics tutor — ask any BIOL2001 Introduction to Quantitative Biology question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.
