Statistics · Ordinary Level
Data & Data Collection
Ordinary Level · Class 1 · Tap NEXT to begin
Section 1 of 2
Types of Data
Statistics is collecting, showing and summarising data (information). First, know your data.
Numerical vs categorical
Numerical = numbers: discrete (certain values, e.g. number of children) or continuous (on a scale, e.g. height). Categorical = groups: ordinal (in order, e.g. grades) or nominal (no order, e.g. gender).
One item or two
Univariate = one item (e.g. number of children). Bivariate (paired) = two items together (e.g. age and weight).
Population & sample
A population is everything that can take part (measuring it all is a census — a true reflection). A sample is a part of it (quicker, cheaper). Sample mean $= \bar{x}$; population mean $= \mu$.
Section 2 of 2
Collecting Data
Primary vs secondary
Primary data = data you collect yourself (advantage: you trust it; disadvantage: time & cost; e.g. a survey). Secondary data = data from another source (advantage: quick; disadvantage: trust; e.g. Google).
Ways to survey
Personal: trustworthy but expensive. Internet/email: cheap but people can lie. Phone: trustworthy but many don’t answer. Observation: trustworthy but expensive. An experiment repeats a process again and again (e.g. counting car colours).
Questionnaires
A set of questions. They must be clear, short, provide options where possible, and avoid leading and personal questions.
Avoiding bias
A good survey must avoid bias in where you ask and who you ask. The sample must be big enough and represent the population — and remember people sometimes lie (e.g. Brexit polls).
You try
Classify: (i) exam grade (A, B, C…); (ii) number of children; (iii) height of a student.
Numbers or groups? Counts or a scale? In order or not?
(i) categorical, ordinal
(ii) numerical, discrete
(iii) numerical, continuous
ordinal; discrete; continuous
That’s Class 1.
Types of data, population vs sample, and how data is collected. Class 2: the three averages.