weeks 1 - 5 quiz 1

What is the mean?

The average value, calculated by adding all values and dividing by the number of values.

What is the median?

The middle value after the values have been ordered.

How is the median found when there is an even number of values?

Take the mean of the two middle values.

What is the mode?

The most frequently occurring value.

What does the horizontal axis usually represent?

The variable positioned along the left-to-right dimension of the graph.

What does the vertical axis usually represent?

The variable positioned along the bottom-to-top dimension of the graph.

What is an independent variable?

The variable manipulated or measured to explore its relationship with the dependent variable.

What is a dependent variable?

The outcome variable measured to see how it is related to or affected by the independent variable.

How can you identify the dependent variable in a graph?

Ask which variable is the outcome being explained or predicted.

How can you identify an independent variable in a graph?

Ask which variable is being used to explain, predict, group, or compare the outcome.

What does a correlation coefficient describe?

The direction and strength of the relationship between two variables.

What values can a correlation coefficient take?

Values from negative one to positive one.

What does a positive correlation mean?

Higher values of one variable tend to occur with higher values of the other variable.

What does a negative correlation mean?

Higher values of one variable tend to occur with lower values of the other variable.

What does a correlation near zero mean?

There is little or no linear relationship between the variables.

What does a correlation near positive one mean?

A very strong positive linear relationship.

What does a correlation near negative one mean?

A very strong negative linear relationship.

How is correlation strength determined?

By the magnitude of the correlation, meaning how far it is from zero regardless of its sign.

Are correlations of positive 0.8 and negative 0.8 equally strong?

Yes. They have equal strength but opposite directions.

Why are graphs useful for large datasets?

Humans struggle to recognise patterns and structure when data are presented only as raw values or tables.

What are the main benefits of graphs?

They simplify information, highlight important ideas, reveal patterns, and communicate information quickly.

What is an unavoidable cost of graphing data?

Some information is lost when complex data are simplified.

What terms are treated as equivalent in PSYC2063?

Graphs, charts, figures, and data visualisations.

What is a dataset?

The complete collection of related data being examined.

What is a variable?

A characteristic that can take different values across observations.

What is a value?

A particular measurement or category recorded for a variable.

What is a data point?

An individual observation represented in a dataset or graph.

What is graph content?

The marks that directly represent the data, such as bars, lines, or points.

What is graph scaffolding?

Elements that help the reader interpret the content, such as titles, axes, labels, legends, gridlines, and sources.

What is numerical data?

Data that represent an amount of something.

What is discrete numerical data?

Numerical data that are counted rather than continuously measured.

What are examples of discrete data?

Number of pets, shoe size, or the result of rolling a die.

What is continuous numerical data?

Numerical data that are measured and can take values between scale points.

What are examples of continuous data?

Height, speed, and length.

What is categorical data?

Data in which items are assigned to labelled groups or categories.

What is nominal categorical data?

Categories with no meaningful order that can be rearranged without changing their meaning.

What are examples of nominal data?

Country, tutor name, and postcode when the numbers are used only as labels.

What is ordinal categorical data?

Categories with a meaningful order but without necessarily equal distances between categories.

What are examples of ordinal data?

Grades, medal rankings, and sizes such as small, medium, and large.

What is a visual encoding?

A property of a graph that changes according to the data.

When is colour a visual encoding?

Only when the colour changes according to the data. A single decorative colour is not an encoding.

What does length or height encode?

A numerical value through the proportional length or height of a mark, such as a bar.

What does position encode?

A value or category through where a mark is placed along an axis or spatial scale.

What does area encode?

A value through the total area of a shape, such as a bubble or map region.

What should be proportional to the data in a bubble graph?

The bubble's area, not simply its radius or diameter.

What does angle encode in a pie chart?

The proportion of the whole represented by each segment.

What does the angle or slope of a line segment encode?

The rate and direction of change between data points.

What does colour hue encode?

Different categories using distinct colours, such as red for one group and grey for another.

What does colour shade encode?

Numerical magnitude using different intensities of the same colour.

What does shape encode?

Different categories using different marker shapes.

What does width or thickness encode?

Magnitude through the thickness of a line or flow.

Why must you identify every visual encoding?

A graph may represent several variables at once using different properties, and missing one can change the interpretation.

What is the usual axis rule for bar graphs?

The numerical axis should usually begin at zero because bar length emphasises the magnitude of differences.

Can a bar graph ever use a non-zero baseline?

Yes, but only in rare cases where another baseline is clearly more sensible and the choice is not misleading.

What is the usual axis rule for line graphs?

The numerical axis often does not need to begin at zero because line graphs mainly emphasise change and trends.

What does consistent axes mean?

Graphs intended for comparison should use the same axis ranges and scales.

Why are inconsistent axes across panels problematic?

They can make similar patterns look different or different patterns look similar.

What does complete data mean in graph design?

Available scale points or relevant observations should not be selectively omitted.

What are consistent scale intervals?

Equal visual distances on an axis represent equal numerical differences.

What are consistent bin sizes?

Grouped ranges cover equal intervals so their frequencies can be compared fairly.

What is the principle of proportional ink?

The amount of shaded area used to represent a value should be directly proportional to that value.

Why can oversized icons or two-dimensional shapes be misleading?

Their area may increase much faster than the numerical value they are meant to represent.

What does the principle that graphs only show the data mean?

A graph directly supports only claims about the variables actually measured and displayed.

Why must observations and explanations be separated?

A graph can show a pattern, but several different processes may explain that pattern.

What should you do after identifying a pattern in a graph?

Generate possible explanations, then seek additional evidence that could distinguish between them.

Why can total counts be misleading when groups differ in size?

Larger groups may have larger totals simply because they contain more people, so rates or appropriate denominators may be needed.

Why can the homicide-count graph not establish that large cities are more dangerous?

It shows total homicides rather than homicide rates adjusted for population size.

What did the eclipse and eye-pain maps demonstrate?

Similar spatial patterns can suggest a plausible explanation, but the maps still do not directly prove the individual-level mechanism.

What did the Taylor Swift swear-word graph directly show?

Later albums contained more recorded swear words than earlier albums.

What could the Taylor Swift graph not establish by itself?

Why the number increased or whether it reflected a change in character or quality.

What were alternative explanations for increased swear words across albums?

Changing social norms, a maturing artist or audience, greater creative freedom, more songs, reduced reliance on radio, or audience demand.

What question guided the Wine Winners tutorial?

Whether the outcomes of wine competitions genuinely reflect the quality of the wines.

In the wine-award figure, what did the notation gold to no award mean?

A wine received a gold medal in at least one competition and no award in at least one other competition.

What percentage of repeatedly entered wines received both a gold medal and no award?

Thirty-nine percent.

What did the distribution of wine awards suggest?

The same wine could receive very different awards across competitions, indicating substantial inconsistency.

How strong were correlations between different wine competitions?

They were generally weak, ranging from approximately negative 0.02 to positive 0.33 in the displayed table.

What does a weak correlation between wine competitions mean?

Wines performing well in one competition did not consistently perform well in another.

What did the predicted and observed gold-medal graph compare?

Gold-medal frequencies expected by chance with the frequencies actually observed across five competitions.

How did observed gold-medal frequencies compare with chance predictions?

They were broadly similar, suggesting that repeated gold-medal outcomes were not much more consistent than chance would predict.

What is the main conclusion from the wine competition evidence?

Competition awards were inconsistent and provide weak evidence of a stable underlying wine quality.

Why should a single wine medal be interpreted cautiously?

The same wine could receive a very different result when judged in another competition.

Why is simply staring at a complex graph ineffective?

Its meaning may not become clear without a structured method for decoding its variables and visual encodings.

What is Step 1 of the six-step graph-reading process?

Read the title, caption, description, and source, then consider the source's credibility.

Why is the caption important?

It may define the population, exclusions, measures, calculations, and conditions needed to interpret the graph correctly.

How should a graph with no title, caption, or source be treated?

It can be described cautiously, but its origin should be verified before using it to support strong decisions or claims.

What is Step 2 of the graph-reading process?

Identify the axes, units, minimum and maximum values, and the meaning of any legend.

Why should you check axis minimums and maximums?

They reveal the scale of the data and whether visual differences may be compressed or exaggerated.

What is Step 3 of the graph-reading process?

Identify each visual encoding and state which variable it represents.

What is Step 4 of the graph-reading process?

Choose one or more data points and explain exactly what each one means.

Why should you orient yourself using individual data points?

It confirms that you understand how the variables and encodings combine before interpreting the full pattern.

What is Step 5 of the graph-reading process?

Read any annotations because they usually highlight patterns or observations the graph creator considers important.

What is Step 6 of the graph-reading process?

Zoom out and identify overall patterns, trends, and relationships between the variables.

What should you list before describing an overall graph pattern?

The dependent variable and every independent or grouping variable represented in the graph.

What is a main effect in graph interpretation?

The overall relationship between one independent variable and the dependent variable while averaging across other variables.

What does it mean to collapse across a variable?

To mentally average across its levels because it is not the variable currently being examined.

How can you visually estimate a main effect across grouped bars?

Compare the average height of the bars for each level of the variable of interest.

What is an interaction?

The effect of one independent variable on the dependent variable changes depending on another independent variable.

What is one way to describe an interaction?

Ask whether the effect of Variable A changes across the levels of Variable B.

What is the equivalent second way to describe an interaction?

Ask whether the effect of Variable B changes across the levels of Variable A.

How can an interaction be identified using slopes?

Compare the slopes for different groups. Clearly non-parallel slopes indicate that the pattern differs between groups.