Descriptive Statistis Flashcards
(27 cards)
What are the three branches of statistics
- Descriptive statistics (organising, summarising and describing data)
- Correlational (exploring relationships)
- Inferential (generalising findings)
State what it is meant by the key term ‘variables’
Variables are measurements made within a study
What are the 4 types of variables (on the chart, in order)
- Organismic (top)
- Environmental (bottom)
- Discrete (left)
- Continuous (right)
Explain what it is meant by organismic (top) variables
Any measurement that can describe a characteristic of an organism
Explain what it is meant by environmental variables
Any measurement to describe an organisms surroundings
State 2 facts to explain what it is meant by discrete variables
- Any parameters that can only take one score
2. Can’t be sub-divided
Explain what it is meant by continuous variables
Measurements that can always be sub-divided more
State, and give an example, of all the combinations that can be made on the variable graph
- Discrete/organismic - biological sex
- Discrete/environmental - treatment used in the experiment
- Continuous/organismic - body weight
- Continuous/environmental - temperature
State, from bottom to top, the levels of measurement (LoM)
- Nominal data
- Ordinal data
- Interval data
- Ratio data
Explain nominal data (3 points)
- About identity/categories (eg - male vs female)
- Can’t infer order of magnitude
- Mutually exclusive and comprehensive
Explain ordinal data (3 points)
- Can infer order (eg - small to large)
- Don’t know the absolute magnitude of each data sample
- all the aspects of nominal data
Explain interval data (4 points) - example being shoe sizes
- Can include difference between categories
- Know the absolute score of each category
- Can’t say A is x times bigger than B
- everything in interval and nominal data
Explain ratio data (2 points) - example being 100m sprint
- Can express differences between data sets as magnitudes, percentages, fold changes, etc…
- everything in interval, ordinal and interval level data
State 2 things should should always/never do when using units
- Always - leave a space between the measurement and the unit
- Never - pluralise (add an s on) or italicise (write in italics)
State 2 positives and 1 negative of using the Mode
- Extreme outliers don’t impact
- Used with any LoM
- It’s a terminal statistic, all it shows is the mode
State 2 positives and one negative of using the Median
- Extreme outliers have no effect
- Any LoM above nominal data
- Ignores values - only using the median
State 1 positive and one negative of using the Mean
- Takes into account and the weight of each measurement
2. Can only be used with interval and ratio LoM’s
State what it is meant by the key term ‘asymptotic’ when talking about distribution
If a distribution curve is asymptotic, it means that the distribution curve never hits zero
State what it is meant by ‘symmetrical’ when talking about distribution curves
If a distribution curve is symmetrical, then it means the mode, median and the mean are all in the same place on the curve
State what it is meant by the ‘point of inflation’ on a distribution curve
- Where curve changes from convex to concave
- Shows the SD
- 80% should be within 1SD of the mean
State what it is meant by the key term ‘Kurtosis’
Kurtosis refers to how we size of the SD: mesokurtic, leptokurtic and platykurtic
Explain the three types of Kurtosis
- Mesokurtic - usual SD
- Leptokurtic - very thin SD
- Platykurtic - very wide/varied SD
State what it is meant by the key term ‘Z score’
The Z score is the number of SD’s you are away from the mean
What are the two types of skew
- Positive skew - on the right of the curve
2. Negative skew - on the left of the curve
