MEDD8815 Introduction to Statistical Methods

How MEDD8815 is assessed

The course outline lists these three components and they total 100%. It also states that homework is due at the meeting after it is set, except the last assignment, which is due the week after the final session; that no late homework assignment will be accepted; and that quizzes cannot be made up. Letter grades are derived from the course score on the scale published in the outline, from A+ at 96 and above to F at 59 and below. Confirm deadlines, submission settings and any change to the assessment on the course Moodle page.

Current MEDD8815 dates

Dates are as published in the submission points on the course Moodle page for this offering. Confirm exact deadlines and submission settings in the live LMS.

What MEDD8815 covers

Eight chapters follow the eight published sessions, from describing one variable and the relationship between two, through producing data and the logic of inference, to t procedures, proportions and two-way tables, regression inference and analysis of variance.

The course outline states that no prior knowledge of statistics is required, although arithmetic and basic algebra are used throughout, and its aim is basic statistical literacy: carrying out standard analyses yourself and reading journal articles that use them. It also prepares students for later courses in regression, experimental design and measurement.

Analyses are run in SPSS, and the set textbook is the eighth edition of Introduction to the Practice of Statistics by Moore, McCabe and Craig.

Distributions, Summary Statistics and the Normal Curve

The first session describes a single variable.

It sorts variables by measurement level, matches each to a graph, and builds the paired summaries of centre and spread: the mean with the standard deviation for symmetric data and the median with the interquartile range for skewed data.

The 1.5 times IQR rule flags suspected outliers, and the session ends with density curves and Normal calculations using Table A in both directions.

Scatterplots, Correlation and Least-Squares Regression

The second session moves to two quantitative variables.

Scatterplots are read for direction, form and strength, correlation measures the strength of a straight-line pattern, and the least-squares line summarises it for prediction.

The session is careful about limits: residuals, influential points, extrapolation, and the three explanations of an association, causation, common response and confounding.

Producing Data, Probability and Random Variables

The third session asks where trustworthy data come from.

It separates observational studies from experiments, sets out control, randomisation and replication, and covers simple random and stratified samples, bias and variability, and parameters against statistics. Probability is introduced as a long-run proportion, with random variables and the law of large numbers.

Sampling Distributions, Confidence Intervals and Tests

The fourth session is the hinge of the course.

It develops the sampling distribution of a mean and of a proportion, the central limit theorem, confidence intervals with their margin of error and sample-size planning, and significance tests with P-values, the two types of error and power, all in the simplest setting of a known standard deviation.

Inference for Means with t Procedures

The fifth session replaces the known standard deviation with the sample one and moves to t distributions.

It covers one-sample procedures, matched pairs analysed through their differences, and two-sample procedures with conservative and software degrees of freedom and the pooled alternative, together with the SPSS dialogs for each.

Inference for Proportions and Two-Way Tables

The sixth session handles categorical responses: intervals and tests for one and two proportions, with the correct standard error for each purpose, and the chi-square test for two-way tables and for goodness of fit, including expected counts, degrees of freedom and the conditions that make the test valid.

Inference for Simple and Multiple Regression

The seventh session attaches inference to the regression line: the model and its conditions, the t test and interval for the slope, the ANOVA table, intervals for a mean response and for a single prediction, and multiple regression with its overall F test and individual t tests.

One-Way and Two-Way Analysis of Variance

The final session compares several means with one-way ANOVA, follows up with planned contrasts or adjusted multiple comparisons, and extends to two-way designs with main effects and interaction.

How to use this guide

Read each chapter before its session for vocabulary, work its examples straight after the session while the homework is fresh, and do the practice cards the evening before the next class, because the quiz at the start of each class covers the previous meeting.

The practice chapter mixes problems across all eight sessions, and the closing chapter is a checklist for the last minutes before a homework deadline and before a quiz.

Evidence and assessment control

Assessment labels and weights follow the course outline for this offering.

Teaching explanations, data sets, worked examples and practice items are independently authored by AskSia and are not University assessment material; any marks shown on them are an AskSia study allocation. Confirm deadlines, submission settings and any change to the assessment on the course Moodle page.

Worked example · free

A confidence interval for a mean, with the software step

Key terms

Distribution

The pattern of values a variable takes and the frequency of each, displayed with a graph and summarised by centre and spread.

Standard Deviation

A measure of spread equal to the square root of the average squared deviation from the mean, computed with divisor n minus 1.

Correlation

A measure of the direction and strength of the straight-line relationship between two quantitative variables, always between minus one and one.

Least-Squares Line

The line that minimises the sum of squared vertical distances from the data points, used to predict a response from an explanatory variable.

Randomised Experiment

A study that assigns treatments to units by chance, the design that allows a difference in responses to be attributed to the treatments.

Sampling Distribution

The distribution of a statistic over all possible samples of the same size, the basis of every confidence interval and test in the course.

Confidence Interval

An estimate plus or minus a margin of error, produced by a method that captures the parameter in a stated proportion of samples.

P-value

The probability, if the null hypothesis were true, of a result at least as extreme as the one observed; smaller values are stronger evidence against it.

Degrees of Freedom

The index of the t, chi-square or F distribution used for a procedure, determined by the sample sizes and the number of estimated quantities.

Chi-Square Test

A test comparing observed counts in a table with the counts expected under a hypothesis, using the sum of scaled squared differences.

Analysis of Variance

A method for testing whether several population means are equal by comparing variation between groups with variation within them.

MEDD8815 FAQ

Questions students ask in each MEDD8815 chapter

When should I report the median instead of the mean?

Use the median together with the interquartile range whenever the distribution is clearly skewed or has outliers, because both are resistant. Use the mean with the standard deviation only for roughly symmetric data without outliers. Plotting the data first is how you tell which case you are in. Chapter 01 · Distributions, Summary Statistics and the Normal Curve

Does a strong correlation mean one variable causes the other?

No. A lurking variable can drive both, which is common response, or two influences can be tangled together, which is confounding. Only a well-designed experiment gives fully convincing evidence of cause; without one, look for strength, consistency, dose-response, time order and a plausible mechanism together. Chapter 02 · Scatterplots, Correlation and Least-Squares Regression

Why is random assignment better than letting people choose their group?

People who choose a treatment tend to differ from those who do not, in motivation, background or prior attainment, and those differences become confounded with the treatment. Random assignment spreads such differences evenly by chance, so a large difference in responses can be credited to the treatment. Chapter 03 · Producing Data, Probability and Random Variables

What does 95 per cent confident actually mean?

It describes the method, not the single interval you computed. If you drew many samples and built an interval from each in the same way, about 95 per cent of those intervals would capture the true parameter. Your one interval either contains it or does not. Chapter 04 · Sampling Distributions, Confidence Intervals and Tests

How do I halve an SPSS P-value correctly?

SPSS prints a two-sided Sig. value. If your alternative is one-sided and the sample result lies in the direction it predicts, divide Sig. by 2. If the sample result lies in the opposite direction, the one-sided P-value is one minus half of Sig., and the data give no support to the alternative. Chapter 05 · Inference for Means with t Procedures

When can I use the large-sample procedures for a proportion?

For an interval, the sample should contain at least 10 successes and 10 failures. For a test, the expected numbers under the null value should both be at least 10. When the counts are smaller, the Normal approximation is unreliable and an exact method is needed. Chapter 06 · Inference for Proportions and Two-Way Tables

Why is the prediction interval wider than the confidence interval for the mean?

The mean response interval only has to account for uncertainty in where the population line lies. A single new case also varies around that line by the full residual standard deviation, which never shrinks however much data you collect, so its interval must be wider. Chapter 07 · Inference for Simple and Multiple Regression

Why not just run a t test for each pair of groups?

Each test carries its own chance of a false alarm, so running many of them makes at least one spurious significant result increasingly likely. ANOVA tests all the means with one F test, and follow-up comparisons use methods that adjust for how many comparisons are made. Chapter 08 · One-Way and Two-Way Analysis of Variance

How to prepare for the assessments

Treat the course as eight short cycles rather than one long one. In each cycle, read the chapter before the session, work its examples by hand and in SPSS straight afterwards, submit the homework well before 6:30 PM on the next Thursday, and review the practice cards the evening before class for the quiz.

Keep one running sheet of the standard error and degrees of freedom for every procedure you meet, because from Session 4 onwards every method is the same reasoning with a different entry on that sheet.

Your AI Statistics tutor for MEDD8815

Stuck on a hard MEDD8815 question? Sia is AskSia’s AI Statistics tutor — ask any MEDD8815 Introduction to Statistical Methods 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.