Index

abbreviations and acronyms for variables 21, 40

accuracy 39

algorithms xvi, 234, 235

clustering 105, 108, 109, 114

alpha level (or significance level) 74–5, 86

alternative hypothesis (H1) 71–2, 74, 75, 76, 83, 86

analytics anxiety 5–11

causes of 7–8

concept of and key components 6

key risks 11

remedies 8–9

analytics biases 12, 125–36

data analysis 128, 131–2, 136

data communication and usage 128, 132–4, 136

data gathering 128, 129–31

key concepts 126

analytics roles 231, 232, 233

analytics software presentations xxii, 189–96

demo demons 189, 191, 192–3

good practices 193–4

analytics software tools 8, 23, 60, 116, 233, 236

analytics trends 229, 233, 234–5

ANOVA (analysis of variance) 68–9, 84

application-related questions 142, 144, 145–6

artificial intelligence (AI), transparent 235

audience(s)

attitude about topic 174

building common ground with 171, 176, 183, 194

call to action 174, 176, 180, 183, 184, 185

involvement 156, 162, 166, 167

triggering emotions in 171, 174, 184

autonomy-supporting language 206, 207, 208

average dispersion see variance

averages 33, 40, 41

see also mean; median; mode

b values 62–3, 83

bad data-based news xxii, 199–214

acceptance stage 203

communication strategies 206–8

resistance as challenge of 206, 207

common types

goal failure 201

insufficient competitiveness 201, 203

negative trends 201

comprehension stage 203–4

communication strategies 204–6

confusion as challenge of 204, 205

dos and don’ts of communicating 212, 213

as game changer 203

motivation (action stage) 204

communication strategies 209, 211

frustration as challenge of 209, 211

bar charts 23, 159, 164, 180

beta values (b) 63, 83

biases see analytics biases

big data xvi, 9, 109, 115, 235

big knowledge 9

bimodal distribution 28

binary variables 19, 20, 29, 33, 55

biserial correlation coefficient 55

bivariate data 46, 48, 84

black box allergy 8

categorical variables 18–19, 20, 28, 29, 33, 65–8

binary 19, 20, 29, 33, 55

nominal 19, 20, 33, 55

ordinal 19, 20, 29, 33, 54

causation, correlation vs 55, 83, 133

causation bias 133

cause-and-effect relationships 76, 80

central location see central tendency

central tendency, measures of 27–33, 34

chart shock 8

charts

DESIGN principles xxi, 156–68

Decluttering 156, 158, 166, 167

Emphasising 156, 158–9, 166, 167

Storifying 156, 160, 161, 166, 167

Involving the audience 156, 162, 166, 167

Giving meaning 156, 163–4, 167

No distortion rule 156, 164, 166, 167

labelling 158, 163, 166

symbols 163, 166

titles or captions 163, 166

see also bar charts; dendrograms; graphs; histograms; line charts; maps; pie charts; scatter plots

cloud analytics 234

cluster analysis xxi, 105–21

hierarchical 109, 113–14, 115, 117, 120

how it works 108–10

k-means clustering 109, 120

key concepts 106

limitations of 117–18

outliers and 115, 117

risks 120

similarity measures 110–13

visualisations 113–16

cognitive computing 234

Cohen’s d 76

collaborative analytics 234

comparison

chart format to enable 159

of group means 66–8, 76

competitiveness, insufficient 201, 203

complex relationships xx–xxi, 91–102

conceptual models 100

confirmation bias 130–1

confounding variables 81–2, 83, 86, 131

confusion 204, 205

continuous variables 18, 19–20, 28, 33, 54, 55, 64

interval 19, 20, 29, 30, 33, 54, 64

ratio 19, 20, 29, 30, 33, 54, 64

sizes of relationships between 76

controlling language 206–8

convenience sampling 79

conversations xvi–xvii

correlation 49–50, 52–5, 84

direction and strength of see correlation coefficient

negative linear 50, 52, 53, 85

no correlation 50

positive linear 49–50, 52, 53, 84

scatter plots 48–50, 51

vs causation 55, 83, 133

correlation coefficient 52–5, 60, 83

biserial 55

Cramer’s V 55

Kendall’s tau 54

Pearson 52–4, 76

point-serial 55

Spearman’s rho 54

counter-arguing 208

critical mindset 229, 231, 233

curse of knowledge bias 132–3, 134, 193, 194

curvilinear relationships 50–1

dashboards

making data meaningful in 163

storified 160, 161, 178–80, 182

data analysis 142

asking questions related to 142, 145

biases 128, 131–2, 136

rigour and relevance of 144

Data Central 233

data communication and usage biases 128, 132–4, 136

data disagreements xxii, 217–26

data distortions xxi

avoiding, and chart design 156, 164, 166, 167

see also analytics biases

data fatigue 8

data fluency xvi, xxii, xviii–xx

elements for sustained

critical attitude 229, 231, 233

monitoring trends 229, 233, 234–5

resources 229, 231, 233

understanding and skills 229, 230

five step action plan 236

data gathering biases 128, 129–31

data quality 142, 144–5

data quality paranoia 8

data scepticism 8

Data Science Central 231

data screening 163–4

data sources 142, 144–5

data storytelling xxi–xxii, 8, 9, 160, 171–86, 230

audience

attitude about topic 174

building common ground with 171, 176, 183

call to action and rewards 174, 176, 180, 183, 184, 185

triggering emotions in 171, 174, 184

canvas 172, 174–8, 181

dashboard example 178–80, 182

moral of the story 180, 184

personal style in 184–5

situation–complication–resolution format 176, 180, 183–4, 185

specific–explore–generalise format 176, 177

surprise story approach 176–7

use of voice in 171, 177–8, 184–5

see also storifying charts

data visualisation 153–68, 172, 230

cluster analysis 113–16

complex relationships 100–1

and data disagreement 220–1, 222

frequency distributions 23, 25

tools 23, 60, 116, 236

see also charts

decimals 39, 40, 41

decluttering 156, 158, 166, 167

demo demons 189, 191, 192–3

dendrograms 109, 114–16, 121

dependent variables 21, 22, 40, 81

deviation 35–8

chart format to emphasise 159

direction and strength of relationship see Pearson correlation coefficient

discrete variables 19–20

dispersion see spread

distance matrix 114

distance metrics see Euclidean distance

distortions see data distortions

distributed analytics 233, 234

distribution see frequency distributions; normal distributions

Drucker, Peter 147

Dunning–Kruger effect 133

edge analytics 233, 234

effect sizes 76–8, 86

emotion(s) 160, 185

triggering 171, 174, 184

emphasising, and chart design 156, 158–9, 166, 167

error of the model (residual) 57

eta squared 76

Euclidean distance 110, 111–13

evidence-based decisions 6, 7

Excel 236

experimental hypothesis see alternative hypothesis

experimental research designs 80, 83

exponential relationships 51

factor analysis 117–18

feedback 93, 147, 162

frequency distributions 15, 23–33

concept of and key components 24

measures of central tendency 27–33, 34

pointiness (kurtosis) 25–7, 40

presentation format 23, 24, 25

skewed 25–7, 30–1, 40

frustration 209, 211

full mediation 99

gain-framed perspective 206

Gartner.com 233

Gaussian distributions see normal distributions

generalisations 48, 69–76, 82, 83

goal failure 201

graphic representation see data visualisation

graphs, frequency distributions 23, 24, 25

group means 66–8, 76

group membership, and outcomes, comparing differences between 64–8, 76

grouping see cluster analysis

histograms 23

humour 193, 208

hybrid intelligence 235

hypothesis 71

alternative (experimental) (H1) 71–2, 74, 75, 76, 83, 86

null (H0) 71–2, 73, 74–5, 76, 83, 86

hypothesis testing 71–6

important effects, and significant effects, difference between 76–82

incompetence compensation competence 8

independent variables 21–2, 40, 81

Infoworld.com 233

interaction effect 93

interactive data 162

the intercept 57, 58, 59–60, 83, 85

interquartile range 34–5, 36

Intersection over Union see Jaccard coefficient or index

interval variables 19, 20, 29, 30, 33, 54, 64

involvement, audience 156, 162, 166, 167

Jaccard coefficient or index 110–11

jargon, avoidance of 63–4, 84–6, 176, 204

k-means clustering 109, 120

Kendall’s tau 54

kurtosis 25–7, 40

labelling charts 158, 163, 166

large numbers 39, 40, 41

line of best fit 56–8, 59, 85

line charts 164

linear regression model 56–69

with error term 60, 61

with multiple predictors 61–2

linear relationships 46–50

concept of and key components 47

direction and strength of 51–5

see also correlation; correlation coefficient

LinkedIn 231, 233

loss-framed perspective 206

maps 159

similarity 115, 116, 120

market segmentation 107

mean 27, 29–32, 33, 40

group 66–8, 76

spread of data around 34–8

meaning, giving, and chart design 156, 163–4, 166, 167

median 27, 28–9, 30, 31–2, 33, 40

mediation xxi, 91, 93, 94, 98–9, 100, 101

full 99

partial 99

mediator variable 98, 99, 100, 101, 102

meetups 233

Microsoft Excel 236

Microsoft Power BI 8, 116, 236

mobile analytics 235

mode 27–8, 33, 40

moderation xxi, 91, 93–8, 100, 101

moderator variable 93, 95, 96, 97, 100, 101, 102

multidimensional scaling plot 115

multimodal distribution 28

nasty questions 208–9

negative linear correlation 50, 52, 53, 85

negative trends 201

nominal variables 19, 20, 33, 55

non-probability sampling 79

normal distributions 25, 26

deviations from see kurtosis; skewed distributions

normality bias 132

null hypothesis (H0) 71–2, 73, 74–5, 76, 83, 86

odds ratio 76

ordinal variables 19, 20, 29, 33, 54

outcome variables 21–2, 40, 48, 57, 58–9, 76, 82

see also predictor variables, and outcomes

outliers 16, 30, 31–2, 34, 40, 163

cluster analysis 115, 117

and data analysis 131, 136

overfitting 132

p-values 74–5, 76–8, 86

partial mediation 99

Pearson correlation coefficient 52–4, 76

pie charts 164, 180

point-biserial correlation coefficient 55

population(s) 70–1, 85

positive linear correlation 49–50, 52, 53, 84

Power BI 8, 116, 236

precision 39

predictions 48, 55–69, 82, 83, 85

predictor variables 21–2, 40, 48, 76, 82

and outcomes 57, 58–9, 61–2

mediation analysis 91, 93, 94, 98–9, 100, 101

moderation analysis 91, 93–8, 100, 101

principal component analysis (PCA) 118

probability sampling 79

Python (programming software) xvi, 236

qualitative data 17

quality of data see data quality

quantitative data 17

quantum computing 233, 235

quasi-experimental research 80–1

questions about data (Q&A sessions) xxi, 141–50

analysis-related questions 142, 145

application-related questions 142, 144, 145–6

data source and quality questions 142, 144–5

nasty questions 208–9

risks 149

tenacity 146, 147

timing of 146–7

tone 146

R (programming software) xvi, 236

random assignment/randomisation 80, 81, 86

random sampling 79, 86

range 34–5

interquartile 34–5, 36

ranking, chart format to enable 159

RapidMiner software 8, 236

ratio variables 19, 20, 29, 30, 33, 54, 64

recency effect 211

regression analysis 56–69, 84, 85

regression line (line of best fit) 56–8, 59, 85

regression plane 61–2

regression slope (or gradient) 57–9, 85

representative sample 78–9, 83, 86

research design 78, 83

experimental 80, 83

quasi-experimental 80–1

residual (error of the model) 57

resistance to bad news 206, 207

resources, for sustained data fluency 229, 231, 233

reverse mentoring 8, 231

risk-based segmentation 107

S-shaped relationships 51

sample(s) 70–1, 86

representative 78–9, 83, 86

sampling methods 79

SAS software 8, 236

scatter plots (or scatter diagrams) 48–50, 51, 159

segmentation see cluster analysis

selection bias 130

self-service analytics (or business intelligence) 233, 235

signal-to-noise ratio 71–5

significance level (or alpha level) 74–5, 86

significant effects, and important effects, difference between 76–82

silhouette scores 117

similarity 105, 110–13

measures 120

Euclidean distance 111–13

Jaccard coefficient or index 110–11

similarity maps 115, 116, 120

similarity matrix 114

Simon, Herbert 105

skewed distributions 25–7, 30–1, 40

slope see regression slope

Smedley, Ralph C. xviii

snowball sampling 79

software see analytics software

Spearman’s rho 54

Spinoza, Baruch xviii

spread (dispersion) 34–8

SPSS 236

standard deviation 37–8

stat phobia 7

statistics 15–41, 230

avoiding jargon when talking about 63–4, 84–6

storifying charts 156, 160, 161, 166, 167

storytelling see data storytelling

straight-line relationships see linear relationships

sum of squared errors (SS) 37, 38

survivor bias 130

symbols 163, 166

systematic vs unsystematic variation 71–4

Tableau 8, 23, 60, 116, 236

tables, frequency distributions 23, 25

tails of a distribution 26

taxonomies 105, 107

test statistics 71–5

titles, charts 163, 166

transparent AI 235

trends

in analytics 229, 233, 234–5

chart format to show 159

negative 201

U-shaped relationships 50–1

univariate data 46, 84

use of data, biases in 132–4

variables 17–23

abbreviations and acronyms 21, 40

confounding 81–2, 83, 86, 131

dependent 21, 22, 40, 81

discrete 19–20

independent 21–2, 40, 81

and measures of central tendency 28, 29, 30, 33

mediator 98, 99, 100, 101, 102

moderator 93, 95, 96, 97, 100, 101, 102

see also categorical variables; continuous variables; outcome variables; predictor variables

variables, relationship between 21–2

linear 46–50

concept of and key components 47

direction and strength 51–5

non-linear

exponential 51

S-shaped 51

U-shaped (or curvilinear) 50–1

see also complex relationships; correlation; correlation coefficient

variance 37–8, 41

ANOVA (analysis of variance) 68–9, 84

variation, systematic vs unsystematic 71–4

visualisation see data visualisation

voice, and data storytelling 171, 177–8, 184–5

voluntary response sampling 79