Economic Data Analysis (ECO1017)

• What is one example of where data was used in policy?

  New Jersey in 1992 - increased the minimum wage from $4.25 to $5.05 increasing employment by 13% in New Jersey relative to Pennsylvania
• What is one example of data being used in business?

  Amazon's recommendation system - 35% of Amazon's total sales are through this.
• What is a variable?


  Anything that can be measured and can differ across entities or across time.
• What is a categorical variable?


  A type of variable that can take on one of a limited, fixed number of values, which represent distinct categories or groups.
• What is an example of a categorical variable?


  Colours
• What is a numerical variable?


  A measurable variable in which arithmetic operations are applicable.
• What is an ordinal variable?


  A type of categorical variable that has a clear ordered relationship between its categories, meaning the categories can be ranked or sorted.
• What is an example of an ordinal variable?


  Education level
• What is an important aspect of categories?


  They can be numbered, but they're not numerical values.
• What is an important aspect of ordinal variables?


  Their values can be numbered but they are not usually numerical variables.
• What is the value of the variable?


  The possible outcomes or potential values that a variable can take based on its definition.
• What is the realised (observed) value?


  The actual value that the variable takes after the experiment or observation is made; it's the concrete outcome from a specific trial or data point.
• What is the preferred condition to use a pie chart?


  When there are around 3-5 categories.
• What can the shape of a histogram tell us?


  The probability distribution.
• What is a population?

  A well-defined collection of units to which we want to generalise a set of findings or a statistical model.
• What is a sample?


  A much smaller collection of units derived form a population used to determine truths about that target population.
• What are the two characteristics that make a sample good?

  • Representative - sample includes only members of the population being studied.
  • Random - every member of the population studied has an equal chance of being selected for the sample which prevents bias.
• What are the three common measures of centre?

  • mean - average value
  • median - middle value
  • mode - most common value
• What is the best use case for mode?


  For categorical or discrete data
• What is the best use case for mean?


  For data that is symmetrically distributed without outliers.
• What is the best use case for median?


  For skewed distributions when outliers are present.
• What can the spread/variability of a distribution be described using?


  The percentiles
• What is the five-number summary?

  • Minimum value
  • Q1
  • Median
  • Q3
  • Maximum value
• What are the ranges for outliers?

  • 1.5 x IQR < Q1
  • 1.5 x IQR > Q3
• Why may outliers appear?

  • Measurement errors
  • Data entry errors
  • Sampling errors
  • Natural variation (genuine rare events)
  • Skewed distribution
  • Unexpected external factors
• What are the two measures of spread?

  • Standard deviation
  • Variance
• What does the Standard Deviation show?


  The spread of observations around the mean.
• What does the size of the value of the standard deviation and variance mean?

  • the standard deviation and variance will be large if the observations are widely spread about the mean and small if the observations are close to the mean.
• Why do we use n - 1 instead of n in the formula for S.D.?


  We use it to account for the constraint in data as the last deviation is already determined as the sum of deviations is zero.
• What does a positive skew look like?


  Tail of the distribution stretches longer to the right (towards higher values)
• What is the order of the measures of centre in a positive skew?


  Mean > Median > Mode as the mean is pulled up by the higher values in the tail.
• What does a negative skew look like?


  Tail of the distribution stretches longer to the left (towards lower values)
• What is the order of the measures of centre in a negative skew?


  Mean < Median < Mode as the mean is pulled down by the lower values in the tail.