MATH2089 Chap.8 Probability and Descriptive Statistics
Probability and Descriptive Statistics
Statistical analysis begins with how observations were produced. This chapter distinguishes population from sample and parameter from statistic, then connects random sampling with the scope of inference. It compares graphical and numerical summaries for centre, spread, shape and unusual observations, and reviews event rules, complements, conditional probability and independence.
The main discipline is to keep a precise sample description separate from a broader population claim. A population is the full group or process of interest, while a sample is the set actually observed. A parameter describes the population and a statistic is calculated from the sample. Keeping those nouns attached to their targets prevents a sample mean from being reported as if it were the unknown population mean.
It also makes clear what later intervals and tests are trying to learn. Sampling design controls the reach of inference. Random sampling can support generalisation to the population represented by the frame. Random assignment can support a causal comparison between treatments. A convenience sample may still be described accurately, but precise arithmetic cannot repair selection bias.
State the observational unit, selection mechanism and important missing groups before making a broad claim. Descriptive summaries work as a set. Mean and standard deviation are useful for a roughly symmetric distribution without dominating extremes. Median and interquartile range resist a small number of extreme values and are often more informative for skewed data.
Range, variance and standard deviation measure different aspects of spread. A complete description covers centre, spread, shape and unusual observations rather than listing every available statistic. Choose a graph to match the variable and question.
Bar charts compare category counts, histograms display a quantitative distribution across intervals, boxplots condense ordered positions and scatterplots show relationships between two quantitative variables. Axes, units and bin choices affect the apparent pattern. Inspect the data context before treating an outlier as an error; it may be a genuine rare case or evidence of another process.
Probability language converts events into operations. Complements handle at-least-one questions, intersections represent simultaneous events and conditional probability changes the reference set after information is supplied. Mutually exclusive events cannot occur together. Independent events do not change one another's probabilities.
Events with positive probability that are mutually exclusive are therefore generally not independent. Translate words into a named event before substituting a formula. End with a claim that matches the evidence: describe the observed sample, identify the design features that permit inference, and keep association separate from causation unless assignment supports the causal step.
What this chapter covers
- 01
Populations, samples and parameters
- 02
Random sampling and bias
- 03
Mean, median and quantiles
- 04
Variance and standard deviation
- 05
Histograms, boxplots and scatterplots
- 06
Conditional probability and independence
Worked example · free
Compare mean and median
- +1The sum is 40, so the mean is 8.
- +1The ordered middle value is 5, so the median is 5.
- +1The value 20 pulls the mean upward through its full magnitude.
- +1The median better represents the centre of this right skewed set, while the extreme value should still be reported.
Key terms
- Population parameter
- A fixed numerical feature of the full population of interest.
- Sample statistic
- A numerical summary computed from observed sample data.
- Random sampling
- Selection governed by a probability mechanism that supports sampling uncertainty calculations.
Probability and Descriptive Statistics FAQ
Can a large sample remove selection bias?
No. A large sample reduces random variability within the sampling frame, but it does not repair systematic exclusion, nonresponse or measurement bias. Design determines representativeness.
When is the median preferable?
The median is often more representative when a distribution is strongly skewed or contains a small number of extreme values. Report spread and shape as well as centre.
Are mutually exclusive events independent?
Not when both have positive probability. Mutual exclusivity makes their intersection zero, while independence would require the intersection probability to equal a positive product of probabilities.
Exam move
For each dataset, write the observational unit, variable type, sampling process and target population before calculating. Practise one descriptive paragraph that covers centre, spread, shape and unusual values. Translate probability phrases such as at least, given that and independent into event notation before applying rules.
Build pairs of examples where mean and median tell different stories, and match each with an appropriate spread summary and graph. Change a histogram bin width or axis scale and explain which visual claims remain supported. For probability, draw a two-event diagram and calculate complements, intersections and conditionals from the same values.
Include one pair of mutually exclusive positive-probability events and show algebraically why they are not independent. End each answer with the scope permitted by the design so sample description, population inference and causal language never blur together.
Working through Probability and Descriptive Statistics in MATH2089? Sia is AskSia’s AI Mathematics tutor — ask any MATH2089 Probability and Descriptive Statistics question and get a clear, step-by-step explanation grounded in how MATH2089 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.