Data Analysis

Data Analysis and Statistics Basic Terms

Understand datasets, variables, populations, samples, center, spread, and relationships before selecting analytical methods.

12 matching terms

Data structure

Dataset

Meaning

A structured collection of related observations used for analysis.

When to use it

Define the dataset's scope, period, source, and unit before analysis.

Practical example

Dataset: daily orders from 2025-01-01 through 2025-12-31.

Data structure

Observation

Meaning

One recorded case, event, person, item, or time point in a dataset.

When to use it

Use it to clarify what one row represents.

Practical example

One observation represents one completed order.

Data structure

Variable

Meaning

A measurable attribute that can take different values across observations.

When to use it

Identify variables before selecting summaries, charts, or models.

Practical example

Variables: order_value, region, channel, and order_date.

Data structure

Categorical variable

Meaning

A variable whose values represent groups or labels rather than measured amounts.

When to use it

Use counts, proportions, bar charts, or encoded categories to analyze it.

Practical example

channel = web, store, partner

Data structure

Numerical variable

Meaning

A variable represented by numbers for which arithmetic operations are meaningful.

When to use it

Use distribution, center, spread, correlation, or regression analysis.

Practical example

order_value = 49,900 KRW

Population and sample

Population

Meaning

The complete set of units or events that an analysis aims to describe.

When to use it

State it explicitly so conclusions are not generalized beyond the intended scope.

Practical example

Population: all customers who placed an order in Korea during 2025.

Population and sample

Sample

Meaning

A subset of a population selected for measurement or analysis.

When to use it

Use representative sampling when analyzing the entire population is impractical.

Caution

A large sample can still be biased if the selection process is not representative.

Practical example

Randomly sample 2,000 orders from the full-year population.

Center

Mean

Meaning

The sum of numerical values divided by their count.

When to use it

Use it as a center measure when extreme values do not dominate the distribution.

Practical example

mean_order_value = sum(order_value) / count(order_value)

Center

Median

Meaning

The middle value after observations are ordered.

When to use it

Use it for skewed distributions or data with influential outliers.

Practical example

Report median delivery time when a few delayed orders are extremely late.

Spread

Variance

Meaning

The average squared distance of values from their mean, with the exact denominator depending on context.

When to use it

Use it in statistical modeling and when decomposing sources of variability.

Practical example

sample_variance = sum((x - mean(x))^2) / (n - 1)

Spread

Standard deviation

Meaning

The square root of variance, expressed in the same unit as the data.

When to use it

Use it to communicate typical variation around the mean.

Practical example

Average wait = 8 minutes; standard deviation = 2.1 minutes.

Relationships

Correlation

Meaning

A measure of the direction and strength of association between variables.

When to use it

Use it to screen relationships before deeper causal or predictive analysis.

Caution

Correlation does not by itself establish causation.

Practical example

corr(ad_spend, sales) = 0.72