This is How Easy It Is to Lie With Statistics

The video explores the power and misuse of statistics in various contexts, including marketing, legal cases, and media. It highlights how Target used statistical models to predict pregnancies, leading to privacy concerns. The video also discusses famous court cases where statistical errors led to wrongful convictions. It examines misleading statistics in advertising and the difference between correlation and causation. Finally, it addresses Simpson's paradox and how data representation can be manipulated to mislead audiences.

What was the main goal of Target's statistical algorithm?

• To reduce shopping cart abandonment

• To predict customer pregnancies

• To increase lotion sales

• To track customer loyalty

In the case of Janet Collins, what statistical error was made?

• Assuming correlation implies causation

• Using a biased sample

• Prosecutor's fallacy

• Ignoring outliers

What was the statistical error in Sally Clark's case?

• Misinterpreting correlation

• Using outdated data

• Assuming events were independent

• Ignoring genetic factors

What was misleading about the Colgate advertisement?

• It used false data

• It misrepresented the percentage of dentists

• It claimed Colgate was the only recommended toothpaste

• It showed incorrect product images

What was the impact of the misleading birth control pill statistics in the UK?

• A significant number of women stopped using the pill

• A decrease in teenage pregnancies

• Increased sales of the pill

• An increase in the pill's price

How can correlation be misleading in statistical analysis?

• It is irrelevant in data analysis

• It can be influenced by a third factor

• It is always a result of random chance

• It always indicates causation

What is Simpson's paradox?

• A method to calculate probabilities

• A type of correlation that implies causation

• A statistical error due to small sample size

• A situation where a trend appears in different groups of data but disappears when these groups are combined

What does the third-cause fallacy refer to?

• A situation where one event causes another

• A situation where two events are unrelated

• A situation where two events cause each other

• A situation where two correlated events are caused by a third factor

What is a common issue with bar graphs that do not start at zero?

• They are not colorful enough

• They can exaggerate differences

• They are difficult to read

• They are too complex

How can statistics be used to mislead without using incorrect data?

• By using complex equations

• By presenting data in a misleading way

• By using large sample sizes

• By avoiding statistical tests