Critical Statistics: Seeing Beyond the Headlines

Contents
LIST OF FIGURES AND TABLES
	Figures
	Tables
Abbreviations
Preface
TOUR OF THE BOOK
	Headlines
	Boxes
	Summaries
	Terminology used in this chapter
	‘Seeing Beyond the Headlines ’toolboxes
	Exercises
	Italic, Bold and Underline
GUIDE TO THE WEBSITE
	The Critical Statistics website
ACKNOWLEDGEMENTS
1: 99% of Statistics are made up
	On bullshit
	The world runs on numbers
		Sometimes contraception doesn’t work
	Statistics in the fake news era
	Don’t be part of the problem
		Statistics for social science students
			Numbers and your degree
			Numbers and your career
2: Where do NumbersCome From?
	The wires
	Making the news
	#nofilter
	Where’s the harm?
	A lie can run around the world before the truth can get its boots on
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
		1. WHICH ORGANISATION OR PERSON PRODUCED THE STATISTIC?
		2. IS THE ORGANISATION/PERSON LIKELY TO HAVE AN AGENDA?
		3. DOES THE JOURNALIST CHALLENGE THE STATISTIC?
	Exercises for Chapter 2
		Exercise 1: Follow the press-release
		Exercise 2: Zombie statistics
3: SAMPLES, SAMPLES EVERYWHERE …
	It’s samples all the way down …
		A lot of Australians don’t believe in climate change
		1.9 unemployed people for every vacancy
		Canadians, despite being Canadian, still sometimes kill each other
		Swords and dragons: not just for geeks any more
			Samples almost all the way down …
	Size matters
		Brexit errors
		Percentage points
		Margins of error: the maths bit
		Low fidelity
		Numbers other than percentages
	Bad samples
	Spotting biased samples
		Self-selected samples
		Scientific surveys
		Non-response bias
		Sampling beyond surveys
	The magic of sampling
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
		1. HOW BIG IS THE SAMPLE?
		2. WHAT IS THE RISK OF SAMPLING BIAS?
		3. WHAT IS THE RISK OF NON-RESPONSE BIAS?
		Example
	Exercises for Chapter 3
		Exercise 1: Is it a good sample?
		Exercise 2: Surveying the headlines
4: MEASURE FOR MEASURE
	The dark side of immigration in open, generous Sweden
	Define your terms
		Two is the loneliest number
	How to lie with definitions
	Counting is hard
		Is racism a thing of the past?
	Asking the right questions
		Tracking down the source of the ‘160,000’ figure
		Validity and reliability
	There’s no such thing as a perfect statistic
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
	Exercises for Chapter 4
		Exercise 1
		Exercise 2
5: What Does it Mean to be Average?
	Average man
	The mean doesn’t always mean what you think it means
	Why doesn’t everyone know this already?
	The median: The mean’s under-appreciated brother
		What is a ‘distribution’?
	Averages are not real
		The mode
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
		1. WHAT TYPE OF AVERAGE IS BEING USED?
		2. ARE THERE LIKELY TO BE LARGE OUTLIERS?
		3. IS THE AUTHOR ‘ESSENTIALISING’?
		Example
	Exercises for Chapter 5
		Exercise 1
		Exercise 2
6: Fraction of a Man
	There are two kinds of data in the world
	What’s the point of percentages?
		Per (out of) cent (a hundred)
		Do we need a White Lives Matter?
		Does gun violence mean the USA is deadlier than a warzone?
	Percentages – backwards and forwards
	Risky business
		All about that base
		Likert scales
	Statistics aren’t real
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
		1. IS A COMPARISON BEING MADE IN TERMS OF RAW COUNTS WHEN PERCENTAGES (OR ANOTHER PROPORTION) WOULD BE MORE APPROPRIATE?
		2. IS THE RIGHT PERCENTAGE BEING USED?
		3. IF THE CLAIM IS ABOUT AN INCREASED OR DECREASED RISK, HAS THE BASELINE RISK BEEN PROVIDED?
		Examples
	Exercises for Chapter 6
		Exercise 1
		Exercise 2
7 Cause and Efect
	Kill or cure
	CSI: Causation
		The habits of highly successful people
	The drugs don’t work
		Postscript – fish brain
	‘Have smartphones destroyed a generation?’
		Establishing causation is really hard
	Establishing causation is not impossible
		Natural experiments
		Eliminating alternative explanations using statistics
	The rush to infer causation
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
		1. IS THIS A CAUSAL CLAIM?
		2. WHY MIGHT X AND Y BE ASSOCIATED?
		3. WHICH EXPLANATIONS HAVE BEEN RULED OUT? WHICH EXPLANATIONSREMAIN?
		Example
	Exercises for Chapter 7
		Exercise 1
		Exercise 2
8 Bad Graphics
	Electioneering
	Charts as a collection of ‘visual metaphors’
		A brief history lesson
	Bad charts: A spotter’s guide
		Bar charts
			What are they used for?
			Bar chart or column chart?
			What makes a bar chart misleading?
		Pie charts
			What are they used for?
			What makes a pie chart misleading?
		Line charts
			What are they used for
			What makes a line chart misleading?
		Scatter plots
			What are they used for?
			What makes a scatter plot misleading?
	Nonsense graphs
		Why have one pie when you can have six?
		Obama’s legacy in nine unreadable charts
		The worst graph in the world
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
		1. WHAT TYPE OF CHART AM I LOOKING AT?
		2. WHICH VISUAL ELEMENTS ARE CARRYING INFORMATION?
		3. WHAT QUANTITY DOES EACH VISUAL ELEMENT REPRESENT, AND HOW?
		4. DOES THE CHART ACCURATELY CONVEY THE STORY OF THE DATA?
		Examples
	Exercises for Chapter 8
		Exercise 1a
		Exercise 1b
		Exercise 1c
		Exercise 2
9: Context is Everything
	‘Is that a big number?’
	Four questions
		‘There is no epidemic of police killing black people … ’
			Is the author trying to say that the number is big, or that it is small?
			What contextual information does the author include?
			What potentially relevant information does the author exclude?
			Does the excluded information make the number seem bigger or smaller?
		Emotive statistics
		American carnage
			Is the story trying to say that the number is big, or that it is small?
			What contextual information does the author include?
			What potentially relevant information does the author exclude?
			Does the excluded information make the number seem bigger or smaller?
		The lens of history
		Competing contexts: What is the Queen of England worth?
			What contextual information does the author include?
			What potentially relevant information does the author exclude?
			Does the excluded information make the number seem bigger or smaller?
	Camera tricks
	Summary
	Seeing beyond the headlines
	Exercises for Chapter 9
	Exercise 1
	Exercise 2
10: Do it Yourself
	The gender pay gap
		The data
		How big is the gender pay gap overall?
		Equal pay for equal work?
		Fun with definitions
		Is the gender pay gap caused by sexism?
			It’s not about sexism
			It is about sexism
			The verdict
	Writing up the results
		IMRaD
		How to lie with true statistics
		Cherry-picking
	How to use statistics to tell the truth
	Summary
	Terminology used in this chapter
	Seeing beyond the headlines
		1. HAVE YOU BEEN HONEST WITH YOURSELF ABOUT THE RESULT YOU WANT?
		2. HAVE YOUR METHODOLOGICAL DECISIONS BEEN SELF-SERVING?
		3. HAVE YOU TRIED TO PROVE YOURSELF WRONG?
		Example
	Exercises for Chapter 10
	Seeing beyond the headlines
		1. IS THE NUMBER AN ESTIMATE BASED ON A SAMPLE?
		2. WHAT IS BEING MEASURED?
		3. IS THE NUMBER AN AVERAGE?
		4. IS THE NUMBER A RAW COUNT?
		5. IS THE NUMBER A PERCENTAGE?
		6. IS THE NUMBER A RISK?
		7. IS A CAUSAL CLAIM BEING MADE?
		8. ARE NUMBERS BEING PRESENTED IN THE FORM OF A GRAPH?
		9. IN WHAT CONTEXT HAS THE NUMBER BEEN PLACED?
		10. IS THIS A NUMBER YOU HAVE PRODUCED YOURSELF?
NOTES
	Chapter 1
	Chapter 2
	Chapter 3
	Chapter 4
	Chapter 5
	Chapter 6
	Chapter 7
	Chapter 8
	Chapter 9
	Chapter 10
Index