Intro

Question 1

Definition / equation of churn rate

Answer 1

Cancellations / total subscribers (current + new subscribers)

Question 2

What are the two types of organized observations

Answer 2

Methodology and shape

Question 3

What is the most common shape for data

Answer 3

Table or spreadsheet

Question 4

Variables

Answer 4

The things we measure (columns of a table)

Question 5

Observations / entity / instance

Answer 5

Rows - Individual instances of the things we are measuring

Question 6

Numerical variables

Answer 6

Both the measurement and unit of measurement (without unit a numerical variable is just a number)

Question 7

What are the two ways of getting a number

Answer 7

Counting (discrete) or measuring (continuous)

Question 8

Whole numbers are what type of variable

Answer 8

Discrete variable

Question 9

Partial values are what type of variable

Answer 9

Continuous variable

Question 10

Categorical variables

Answer 10

Characteristics with words or relative values

Question 11

Nominal variable

Answer 11

A categorical variable that is specifically A named value

Question 12

Dichotomous variable

Answer 12

A categorical variable that is binary (yes / no, true / false, on / off, 1 / 0)

Question 13

Ordinal variable

Answer 13

A categorical variable that is a subjective value (a ranking from 1 to 5)

Question 14

3 common messy data problems

Answer 14

1. Typos
2. Missing data
3. Inconsistent coding (three instead of 3 or N/A instead of 0)

Question 15

Missing completely at random

Vs

Missing at random

Vs

Structurally missing

Answer 15

Data was simply not entered or entered properly

We can predict if one value is missing based on the value in another variable

We don’t expect there to be a value to begin with

Question 16

Accuracy

Answer 16

A measure of how well records reflect reality

Question 17

Validity

Answer 17

The data actually measures what we think it is measuring

Question 18

Various ways that a dataset can be low quality

Answer 18

Typos
Mistakes
Missing data
Poor measurement
Duplicate observations

Question 19

What are the two types of categorical variables

Answer 19

Ordinal (ordered)

Nominal (unordered)

Question 20

A distribution is

Answer 20

A function that shows all possible values of a variable and how frequently each value occurs

Question 21

Interquartile range

Answer 21

The range between the first and third quartile of the dataset

First quartile marks the point at 25% into the range of data

Third quartile marks the point at 75% into the range of data

A range of data is all values arranged from smallest to largest

Question 22

Bimodal distribution

Answer 22

A distribution with two peaks (modes)

Question 23

The act of aggregating data

Answer 23

Summarizing a numeric variable across each value of a categorical variable

Question 24

Correlation coefficient

Answer 24

Direction: - or +

Strength: 0 to 1

Question 25

Chart for representing change over time

Answer 25

Line graph

Question 26

Chart for comparing a part to the whole

Answer 26

Pie chart

Question 27

Chart for showing the spread of data points in one variable

Answer 27

Histogram

Question 28

A chart for comparison of two variables to understand a trend

Answer 28

Scatterplot w/ or w/o trendline

Question 29

Univariate charts

Answer 29

Help us visualize a change in only one variable - often that means measuring “how much” - a common type for counts is bar charts

Question 30

Univariate chart type examples

Answer 30

Bar chart
Histogram
Density curve
Box plot
Univariate map

Question 31

Bi / Multivariate charts definition

Answer 31

Charts that show the relationship between two or more variables

Question 32

Multivariate chart examples

Answer 32

Scatterplot
Line chart
Bivariate map

Question 33

Is information redundancy bad?

Answer 33

Not necessarily, redundancy can help add clarity or emphasis

Question 34

Linear scale

Answer 34

The numbers of the axis count up by a consistent interval

Question 35

Logarithmic scale

Answer 35

Common for showing exponential growth

Question 36

Three common color scales

Answer 36

Sequential
Diverging
Categorical

Question 37

Descriptive analysis

Answer 37

We describe, summarize, and visualize data so that patterns can emerge

Most of the time this is the first step in the analysis process

Question 38

Descriptives / summary statistics

Answer 38

Central tendency: mean median mode

Spread: range, quartiles, variance, standard deviation, distribution

Question 39

Exploratory analysis

Answer 39

Typically the next step after descriptive analysis

We look for relationships between variables in our dataset

Question 40

Clustering analysis

Answer 40

Uses Principal Component Analysis which compresses the variables into principle components that can be plotted against each other

The plotting is checked with a k-means clustering value to confirm the correlation

Question 41

Inferential analysis

Answer 41

We test a hypothesis on a sample of a population and then extend our conclusions to the whole population

Typically use an A/B test

Sample size should be at least 10% of the population and must be random selection

Question 42

Casual analysis

Answer 42

Carefully designed experiments, usually with the following:

• only change one variable at a time
• carefully control all other variables
• repeated multiple times with same results

Question 43

Good experimental design

Answer 43

Replication

Randomization

Control

Question 44

Casual inference with observational data

Answer 44

Requires:

• advanced techniques to identify a casual effect
• meeting very strict conditions
• appropriate statistical tests

Question 45

Predictive analysis

Answer 45

Uses supervised machine learning techniques

It will only be as good as the training data used to start the algorithm