PSYC10006 Chap.14 Describing Data and Judging Arguments
Describing Data and Judging Arguments
A column of numbers is almost impossible to interpret directly, and the last five research methods modules start from that fact: data become readable when they are described in terms of how often each value occurs. From a frequency distribution follow three ways of showing one, three answers to what is typical, and three answers to how spread out the values are. The distinctions among them are not arithmetic.
They are about how each measure responds to a value sitting a long way from the rest, which is why the choice of statistic is a decision about the data rather than a matter of preference. The stream then closes on argument itself, with a vocabulary split that decides which words you may use to assess a piece of reasoning.
What this chapter covers
- 01
Why raw data resist interpretation, and what a frequency distribution supplies
- 02
Frequency tables, histograms and boxplots, and what each is best at
- 03
Why looking at the shape first determines every later choice
- 04
Mean, median and mode as three answers to what is typical
- 05
How each measure responds to a value far from the rest
- 06
Range, inter-quartile range and standard deviation, and what each is willing to ignore
- 07
Standard deviation in words: distance, squared, averaged, square-rooted back
- 08
What the subject actually asks of you on the statistics material
- 09
Pairing a measure of centre with a measure of spread that behaves the same way
- 10
Why a number without its companion is not a description
- 11
Deductive and inductive arguments, and the two vocabularies for assessing them
- 12
Why almost every empirical claim in this subject is inductive
Describe an unfamiliar distribution in two numbers
- 1Look at the shape first. A tight cluster with two far-out values is a right-skewed distribution with extremes, and that observation is available before any calculation.
- 1Rule out the mean and say why. Every value contributes its full magnitude, so two times several multiples longer than the rest will drag the mean above anything a typical participant produced.
- 1Choose the median, because it depends on the position of values in an ordered list, so the two long times move it by at most one place and it reports what a typical participant did.
- 1Pair it correctly with the inter-quartile range. Both are position-based, so the pair responds to the data in the same way; a median with a standard deviation would give a robust centre and a spread inflated by exactly the values the centre resists.
- 1Say what the numbers cannot. Note the shape and the two extreme observations explicitly, because a reader who knows the distribution is skewed can interpret two numbers and a reader given only the numbers cannot.
Key terms
- Frequency distribution
- A count of how often each value or band of values occurs in a set of data. It is what makes raw numbers readable, and every description in these modules is built on it.
- Central tendency
- What is most typical in a distribution, measured by the mean, the median or the mode. The three differ in how they respond to values far from the rest rather than in difficulty.
- Inter-quartile range
- The spread of the middle half of the data, ignoring the top and bottom quarters. It describes spread robustly when a distribution has extremes, at the cost of discarding half the observations by design.
- Standard deviation
- A measure of the typical distance between a value and the mean, obtained by squaring those distances so they do not cancel, averaging them, and taking the square root to return to the original units.
- Deductive argument
- An argument claiming that if the premises are true the conclusion must be true. It is assessed as valid or invalid, which are words about structure rather than about how convincing it is.
- Inductive argument
- An argument claiming that if the premises are true the conclusion is probably true. It is assessed as strong or weak, and a strong one can still have a false conclusion.
Describing Data and Judging Arguments FAQ
When should I report the median rather than the mean?
When the distribution is skewed or contains isolated values. The mean is the balance point and every value contributes its full magnitude to it, so extremes pull it away from what is typical. The median depends on position and barely moves. Say which property of the distribution made you choose.
Do I need to be able to calculate a standard deviation?
The stated outcome is to understand how it is calculated, which is comprehension rather than execution, and the module pages give no formula, no dataset and no worked computation. Be able to say what it measures, why the squaring and the rooting are there, and when it is the wrong summary to report.
Why does it matter which measure of spread I pair with which centre?
Because the mean and the standard deviation are both magnitude-based while the median and the inter-quartile range are both position-based. Mixing them reports a spread that is sensitive to exactly the values the centre was chosen to resist, so the two numbers describe the distribution inconsistently.
Why can an inductive argument not be called invalid?
Because induction never claimed the conclusion would follow necessarily, so failing to do so is not a fault in it. Valid and invalid belong to deductive arguments, strong and weak to inductive ones, and an item can test the split by asking which word applies rather than whether the reasoning is good.
What was the whole stream for?
It is a vocabulary for saying how much a piece of evidence supports a claim. That is the skill the written assignment is marked on, the skill half of every quiz tests, and the same skill the neuroscience chapters reached for in distinguishing what imaging can show from what modulation can show.
Assessment move
Work this material on data rather than on definitions. Take any small set of numbers, add one extreme value, and recompute all three measures of centre; watching the mean move while the median and mode stay put teaches the distinction in a way that reading it does not. Then practise the pairing rule until it is automatic, and be able to give the reason, since that is where the explanation mark sits.
On the standard deviation, calibrate your preparation to what is actually asked: be able to explain the four moves in words and what each one is for, rather than memorising notation the subject pages never state. For the final module, drill the two vocabularies as a matching task, and practise assessing an argument in two passes, first asking whether the premises are true and then whether the conclusion follows.
Working through Describing Data and Judging Arguments in PSYC10006? Sia is AskSia’s AI Psychology tutor — ask any PSYC10006 Describing Data and Judging Arguments question and get a clear, step-by-step explanation grounded in how PSYC10006 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
