Tables and graphics

Question 1

Nate Silver

Answer 1

The signal and the noise - 90% of data collected in the last 2 years

Question 2

Charles Babbage

Answer 2

1791-1871
Errors using inadequate data are much
less than those using no data at all

Question 3

John Wilder Tukey

Answer 3

1915-2000
Far better an approximate answer to
the right question, which is often
vague, than an exact answer to the
wrong question, which can always be
made precise

Question 4

Quantitative data reporting through figures

Answer 4

Aims to communicate information without distorting underlying results

Question 5

Three main applications in science

Answer 5

1. Experimental design
2. Exploratory Data Analysis (EDA)
3. Presentation/publication of results

Question 6

Tables

Answer 6

Complex data and actual values.

Don’t tabulate what can easily be said

Question 7

Figures

Answer 7

Trends and patterns and highlighting differences

Question 8

Constructing tables

Answer 8

Always tabulate vertically so you read down, not across.

Question 9

Figure axes

Answer 9

X/horizontal axis = what we manipulate.

Y/Vertical axis = the response

Question 10

Figure variability

Answer 10

Include measure of variation and sample size

Question 11

Exploratory Data Analysis (EDA)

Answer 11

What variation occurs within and between my variables.? Plot data before stats.
Helps to identify underlying structure and decide most parsimonious model.
Identifies outliers.

Question 12

Anscombe Quartet

Answer 12

4 pairs of variables, 11 observations in each, same mean and fitted regression, same R^2 but very different situations.
Everything is not as it seems.

Question 13

Linear model assumptions

Answer 13

1. Linearity between response and predictors
2. Residuals are normally distributed
3. Residuals have equal variances
4. No overly influential points

Question 14

Classes of data

Answer 14

Categorical, ordinal (ranked, ordered), Measurement (ratio (with 0)/interval)

Question 15

Categorical graph type

Answer 15

Bar graph

Question 16

Quantitative graph type

Answer 16

One variable - Box-plot
Two variables - maps
Many - Icon

Question 17

Mixed graph type

Answer 17

Two variables - Bar graph

Many - Bar graph

Question 18

Histograms

Answer 18

For frequency distributions of continuous variables.
Bars drawn together.
X = classes, y= frequency

Question 19

Frequency polygon

Answer 19

Frequency distribution - similar to histogram but with lines

Question 20

Dot plots

Answer 20

Density plot - not affected by subjective choice of number/width of bars. Full dataset.

Question 21

Box-plots

Answer 21

50% of data in the box, 90% of data in the whiskers

Question 22

Notched box-plots

Answer 22

Notch with median and 95% confidence interval

Question 23

Violin plots

Answer 23

Similar to box plots but reveal features - more accurate and reveal true distribution. Width proportional to frequency. Widest point is mode..

Question 24

Pie charts

Answer 24

Rarely used in science - consider stacked bar chart instead

Question 25

Bar graphs

Answer 25

NOT the same as histograms. Bars are separate and represent summary data (mean and variation).
Don’t reveal distribution.

Question 26

Scatter plots

Answer 26

Show degree of association between two continuous variables

Question 27

SPLOM

Answer 27

Scattered plot matrix
Useful for EDA.
Shows shape and distribution of all variables and bivariate relationships between them.

Question 28

Category plots

Answer 28

Scatter plots stratified by a third (categorical or ordinal) variable

Question 29

Line graphs

Answer 29

Time series or temporal data