dab

S in Soar

specify question

O in soar model

obtain the data

A in soar model

analyze the data

R in soar model

report the results

descriptive

what happened?

diagnostic

why did it happen?

predictive

what will likely happen?

prescriptive

what should we do?

Nominal

categories with no order (product color, region, zip code, jersey number)

Ordinal

ordered categories, unequal gaps (Satisfaction, credit rating, t shirt size)

Interval

numerical, equal gaps, no true zero (SAT scores, fahrenheit, calendar years)

Ratio

numerical, equal gaps, true zero (revenue, cogs, age, employee count, units sold)

mean

average

median

middle number when sorted

mode

most frequent number

Positive right skew

mean > median

kurtosis

how heavy tails are

negative left skew

mean < median

Range

max - min

IQR

Q3 - Q1

variance

average squared distance from mean

standard deviation

square root of variance

Z score

value - mean / standard deviation

high standard deviation

data spread out

low standard deviation

data more consistent

five number summary

1. minimum

2. Q1

3. median

4. Q3

5. maximum

what is five number summary equal to horizontally?

box plot

four pillars of data ethics

privacy, transparency, fairness, accountability

benfords law

smaller leading digits appear more frequently

horizontal analysis

current year - previous year/previous year x 100

vertical analysis

individual item/total x 100

breakeven analysis

fixed costs/ price per unit - variable cost per unit

R in AI prompting

role

C in AI prompting

context

T in AI prompting

task

F in AI prompting

format

charts: comparing categories

bar chart

charts: trends over time

line chart

charts: parts of a whole

pie chart

charts: distribution of numerical data

histogram

charts: relationship between two numerical variables

scatterplot

charts: median, spread, outliers

box plot

charts: comparing category composition

stacked bar chart

1 standard deviation

68%, - and + on each side

2 standard deviations

95%, - and + on each side x2

3 standard deviations

99.7%, - and + on each side x3

dimensions tableau

categories (product or region)

measures tableau

numbers (revenue or profit)

tableau dashboard

multiple visuals on one screen

tableau .twbx

packaged tableau workbook

excel csv

plain text, no saved formulas, typically one table, easy to share

excel xlsx

excel workbook, formulas, multiple worksheets

A1

relative, changes when copied

$A$1

absolute, never changes when copied

$A1

column fixed

A$1

row fixed

Referencing same cells across worksheets formula

example: =SUM(Jan:Mar!B2)