PSYB06H3 Applied Statistical Analysis for Psychology & Neuroscience

How PSYB06H3 is assessed

Weights come from the Fall 2026 syllabus grading table and total 100%. The syllabus lists the mock midterm as due October 20, while a later lecture slide says it is due Friday October 23, so check Quercus for the deadline that applies. The final exam date is listed as to be determined. Under the grade reweight rule, a higher final exam grade may override lower grades on the midterm, mock midterm or quizzes; the lecture notes add that no reweight applies to a missed written assignment. Theory prompts require minimum in-person participation, with an alternative assessment for uncontrolled conflicts.

Current PSYB06H3 dates

Dates are as published in the Fall 2026 syllabus and the midterm information page. Confirm exact deadlines and submission settings in the live LMS.

What PSYB06H3 covers

Move from describing a sample to judging how surprising it would be if nothing had changed.

The Fall 2026 syllabus asks students to understand why each test works rather than memorise its equation, because the formulas are derived from foundational mathematics and logic. The comparison it uses is cooking: knowing the ingredients matters less than knowing how to adapt a recipe when the situation changes.

In practice that means every calculation should arrive with a sentence saying what the number tells you about people, brains or behaviour.

A cumulative sequence

The lectures build on each other. Describing a sample comes first, followed by how well a sample represents its population, the rules of probability, and probability distributions such as the normal and binomial curves.

Those pieces combine in hypothesis testing with z and t, then extend to comparisons of two groups using independent and paired samples t tests. After reading week the schedule moves to power, nonparametric equivalents, correlation and regression, chi squared tests, analysis of variance and an introduction to machine learning.

A gap in an early lecture therefore reappears later, which is why the course recommends catching up before the next class.

How the tests are written

Both in-person tests are handwritten and last three hours, and students bring a non-programmable, non-graphing scientific calculator. Questions mix short answers, short calculations and one long calculation built around a data set.

The course-provided formula sheets and tables are handed out, so the challenge is selecting the right procedure, carrying it out accurately and interpreting the result.

The midterm covers Lectures 1 to 6. The final exam focuses on Lectures 7 to 11 with some comprehensive overlap, so the early material keeps earning marks all term.

Using this guide

Chapters 1 to 7 follow the first six lectures in order: measurement and descriptives, sampling and standard error, probability rules, counting and binomial probability, the normal curve, hypothesis testing, and two-sample t tests.

Chapter 8 turns those results into the one-page Results section required by the written assignment. Each chapter pairs a fully worked calculation with practice whose answers follow from printed steps, and the closing chapters rehearse the midterm structure with an original data-set question.

Work with a calculator open, round to four decimal places as the course does, and finish every answer with a plain-language conclusion.

Worked example · free

Locating one score on the normal curve

Key terms

Population Parameter

A value such as μ or σ that describes every member of the population of interest and is written with a Greek letter.

Sample Statistic

A value such as x̄ or s computed from a subset of the population and used to estimate the matching parameter.

Sum of Squares

The total of the squared deviations of each score from the mean, which forms the numerator of the variance.

Degrees of Freedom

The denominator n − 1 used for sample variance, which corrects the tendency of a sample to underestimate population variance.

Standard Error

The standard deviation of a sampling distribution of means, estimated as σ divided by the square root of n.

Central Limit Theorem

The principle that sample means form an increasingly normal distribution around μ as n grows, whatever the population's shape.

Conditional Probability

The probability of one outcome once another outcome is known to have occurred, written p(B | A).

Binomial Probability

The probability of a set number of successes across repeated independent events, combining the multiplicative and additive rules.

Standard Score

A z score that states how many standard deviations a value lies from the mean, so it can be read from the z table.

Null Hypothesis

The default claim that any observed difference is no greater than random noise would produce, which a test can only reject or fail to reject.

Type I Error

Rejecting a null hypothesis that is actually true, which happens at the alpha rate when the null holds.

Statistical Power

The probability of rejecting the null hypothesis when a true effect exists, with 0.80 treated as the minimum to aim for.

Effect Size

A standardised measure of how large a difference is, such as Cohen's d, which does not grow merely because n grows.

Pooled Variance

A weighted average of two sample variances, weighted by their degrees of freedom, used to compute d for two groups.

PSYB06H3 FAQ

Questions students ask in each PSYB06H3 chapter

Why is the mean called the least squared distance value?

The mean is the single point that minimises the sum of squared distances to every score, so it sits as close as possible to all of the data at once. That is why it treats every score equally, and also why a single extreme score can pull it further than it pulls the median. Chapter 01 · Variables, Measurement and Descriptive Statistics

Why does the standard error fall as the sample grows?

The standard error divides σ by the square root of n, so larger samples produce means that cluster more tightly around μ. Intuitively, a mean built from many scores lets high and low values offset each other. Because of the square root, quadrupling the sample is needed to halve the error. Chapter 02 · Distributions, Sampling and Standard Error

How can I tell an AND question from a GIVEN question?

An AND question treats both outcomes as unknown and asks how often they occur together, so it divides by everyone. A GIVEN question tells you one outcome has already happened, so it divides only by that group. Rewording the question as a fraction of a specific group usually reveals which one is meant. Chapter 03 · Probability Rules and Bayesian Reasoning

How do I decide between a permutation and a combination?

Ask whether swapping two selected items creates a different outcome. If assigning the same people to different named roles counts as a new result, order matters and a permutation applies. If only the membership of the group matters, use a combination, which divides out the orderings of the chosen items. Chapter 04 · Counting Outcomes and Binomial Probability

Why can I not find the probability of one exact score on the normal curve?

A single point on a continuous scale has no width, so the area above it is zero and the table has nothing to report. Questions are therefore framed as ranges: above a score, below it or between two scores. A binomial count is different because each discrete outcome has a bar with width. Chapter 05 · The Normal Curve and z Scores

Why can a study never accept the null hypothesis?

Testing assumes the null from the outset and only measures how surprising the data would be under it. A non-significant result may come from insufficient power or a poor design rather than a true absence of effect. The defensible decision is to fail to reject, and any claim that the null is likely true has to be argued separately. Chapter 06 · Hypothesis Testing with z and One-Sample t

How do I decide between an independent and a paired t test?

Ask who produced each score. If every participant appears in only one condition, the groups are independent. If each participant, or each matched pair, contributes one score to both conditions, the design is paired and the analysis uses difference scores. Group sizes are a clue too: paired data always have equal counts once incomplete pairs are removed. Chapter 07 · Independent and Paired Samples t Tests

Will mistakes in my first draft cost me marks?

Mistakes are expected in the first submission. When reviewers judge that a best effort was made, the initial grade is usually 6 out of 10, and the revision can add up to 4 more. Papers that look rushed, omit key statistics or ignore the instructions risk the 4 or 0 brackets instead. Chapter 08 · Writing a Statistical Results Section

How to study for the exam

Revise in lecture order, because each topic assumes the previous one. For every lecture, rebuild one calculation from a blank page with only the course formula sheet beside you, then explain the result in one plain sentence. Practise reading the z and t tables until locating the mean-to-z, larger and smaller portions or a critical t takes seconds.

When a practice answer is wrong, label the fault as the procedure chosen, a formula input, the arithmetic, the table lookup or the interpretation; the label tells you what to drill next. Sit the mock midterm under test conditions, closed book with a calculator, as the course recommends, and time each part against the midterm's mark split.

Keep the written assignment moving alongside calculation practice, since its Results section uses the same tests in the format researchers read. In the final week, prioritise hypothesis testing and two-sample t tests, which combine every earlier skill in a single question.

Your AI Statistics tutor for PSYB06H3

Stuck on a hard PSYB06H3 question? Sia is AskSia’s AI Statistics tutor — ask any PSYB06H3 Applied Statistical Analysis for Psychology & Neuroscience 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.