Knowledge quiz

Identify whether two variables are closely correlated to confidently predict the future

Identify a statistical relationship between two variables, which may (or may not) represent a predictive relationship

Numerically calculate how much of the change in one variable's value is explained by the second variable

Compare the mean values of each variable

Use a histogram.

Always represent in your local currency.

Make sure that the graph line is bold in order to be clear.

Ensure it has a scale in order to avoid giving a misleading impression.

Build a scatterplot to visually evaluate whether a relationship between the variables exists

Build a frequency distribution for each of the variables

Calculate the mean and median of each of the variables

Check for outliers

It works well with distributed, heterogenous data sources.

It only uses proprietary SAP technology.

It should ideally be installed on a mainframe computer.

It specializes in data science projects using artificial intelligence.

0.17 or 17%

0.33 or 33%

0.5 or 50%

0.7 or 70%

Ask for the baseline date. It makes a big difference if that was over the last week, month, year, or decade.

Ask for the absolute number of customers before and after. 2 to 6 customers doesn't sound much. 2000 to 6000 is much more significant.

Check whether the average deal size has gone down or up.

Check whether the customers are in close enough geographies to fulfil the order cheaply and effectively.

Make sure that all the bars have a different color to make the chart interesting for the eye.

If the chart requires more than 5 or 6 bars, break the graph into 2 bar charts.

Use 3D bar charts wherever possible to make the charts look more modern and interesting.

Start the axis at zero to ensure that you don't mislead the viewer with the relative volumes of the bars.

To avoid missing other factors which may explain differences, such as seasonality or organic growth

To show your colleagues that you understand the analysis

To be able to tweak the conclusions to suit your previous beliefs

To understand what variation was added that differs from the control

Distribution curves graph the frequency of variables such as height, weight, test scores, etc.

A distribution curve adds up the frequency of one variable against one or more other variables.

Distribution curves are always in the shape of a bell.

Distribution curves can be used to compare the means of two or more variables to check the significance of their differences.

When you need to describe absolute amounts

When dealing with financial measures

When you are trying to describe volatility rather than absolute amounts

When dealing with time-series data