Instructor mode is active. All missions are navigable and instructor notes appear on each screen.
The computer already calculated the descriptive statistics. Your job is to stay calm, read the evidence carefully, and decide what the numbers allow you to say — Grogu-style patience, Mandalorian discipline.
| Variable | Type | Meaning |
|---|---|---|
| Channel | Categorical | Customer channel: Horeca (Hotel/Restaurant/Café) or Retail. |
| Region | Categorical | Customer region: Lisbon, Oporto, or Other. |
| Fresh | Continuous spending | Annual spending on fresh products. Fresh is the anchor example used to teach each statistic, then compared with the other product categories. |
| Milk | Continuous spending | Annual spending on milk products. |
| Grocery | Continuous spending | Annual spending on grocery products. |
| Frozen | Continuous spending | Annual spending on frozen products. |
| Detergents_Paper | Continuous spending | Annual spending on detergents and paper products. |
| Delicassen | Continuous spending | UCI variable name for annual spending on delicatessen products. |
| Product | Mean | Median | SD | Skewness | Minimum | Maximum |
|---|---|---|---|---|---|---|
| Fresh | 12,000.30 | 8,504 | 12,647.33 | +2.56 | 3 | 112,151 |
| Milk | 5,796.27 | 3,627 | 7,380.38 | +4.05 | 55 | 73,498 |
| Grocery | 7,951.28 | 4,755.5 | 9,503.16 | +3.86 | 3 | 92,780 |
| Frozen | 3,071.93 | 1,526 | 4,854.67 | +5.91 | 25 | 60,869 |
| Detergents / Paper | 2,881.49 | 816.5 | 4,767.85 | +3.63 | 3 | 40,827 |
| Delicatessen | 1,524.87 | 965.5 | 2,820.11 | +11.15 | 3 | 47,943 |
| Row | Channel | Region | Fresh | Milk | Grocery | Frozen | Detergents / Paper | Delicassen |
|---|
Mean is the arithmetic average and is sensitive to extreme values. Median is the middle ordered value and is more resistant to extreme values.
Mode is the most frequent value. For quantitative spending data, a repeated value may exist without being a useful description of a typical customer.
Standard deviation describes spread around the mean in the variable's original units. There is no universal numerical cutoff for “high” or “low”; compare it with the scale and context.
Sample variance is the square of the sample standard deviation. Because it is in squared units, it is usually less intuitive to communicate than standard deviation.
−0.5 to +0.5: approximately symmetric. Between ±0.5 and ±1: moderately skewed. Beyond ±1: substantially/highly skewed. These are interpretation guidelines, not mathematical laws.
This app uses excess kurtosis, where a normal distribution is near 0. Positive values indicate heavier tails. Course screening rule: above +1 means investigate extreme-tail behavior, not automatically delete an observation.
Range = maximum − minimum. It is easy to interpret but depends entirely on two observations, so use it with other measures of spread.
Count gives the number of numeric observations. Sum gives total activity. They provide scale and context, but neither alone describes a typical observation.
SE = s / √n. It describes sampling variability of the sample mean, not the spread of individual customer observations.
Excel-style “Confidence Level (95.0%)” is the margin of error around the sample mean. Mean ± margin produces the confidence interval for the mean.