Introduction to Statistics

Why use statistics

Statistics is a robust way to analyses data in order to make decisions.

• Present the data graphical
• If data is significant different
• If data is has the same shape distribution

Graphs

Control/time chart

• Monitor to production process
• Behavior over time

Correlation Graph

Correlation Graph To find correlations

There is a correlation as the value on the X-axle is having influence on the Y-axle (see A-D)

Histogram

Histogram shape of distribution and position of your data

Data shape and variation

Big Variation

Why is a lot of data normally distributed

Variation of is often caused by multiple factors

Example resistance of a light-bulb
• Length of the wire
• Material of the wire
• Diameter of the wire
• Internal connection of the wire
• External connection of the bulb

Adding up all these distributions will create normally distributed data due to the Central limit theorem.

Central limit theorem

The summarizing independent distributions results in a approximately normally distributed data. Even if the independent distribution are not normally distributed.

Normal assumption

Due the Central limit theorem most data is normally distributed and a lot of statistics assume normality.

What can we do with this?

• Check if data is within tolerance
• Determine if there is a significant difference in Mean or Variation
• If there is a correlation between to data sets