Learn the fundamentals of Data Analytics

Define data analysis as turning a mass of hard-to-understand data into actionable insights, using descriptive statistics to guide decisions across stores, apps, finance, and health.

Explore how data tables store customer, product, and purchase date information to convert data into actionable insights, and how relational and NoSQL databases link them to create a view.

Data analysis turns hard-to-interpret data into actionable insights by summarizing information; descriptive statistics is the branch that summarizes the characteristics of a data set.

Explore the vocabulary of descriptive statistics using an ice cream shop data table, defining population, observation, sample, variables, modalities, and distinguishing qualitative and quantitative, nominal and ordinal, continuous and discrete.

Discover how to use frequency and percentage to summarize data tables, counting observations by gender and rating, and compute cumulative frequencies and percentages for quick insights.

Explore the mode, the most frequent value in a data table for a given variable, illustrated with gender, rating, and ice cream examples, and note its limits.

Compute the mean, the arithmetic average, by summing values and dividing by the number of observations, noting extreme values bias it and that geometric, harmonic, quadratic, and weighted means exist.

Learn how the median provides a central value not biased by extreme values, compare it to the mean, and compute medians for odd and even data sets.

Explore how quartiles partition data at 25%, 50%, and 75% to define Q1, Q2 (median), and Q3, and compare the median with the mean as a central value.

Explore how the range and interquartile range quantify dispersion in data by comparing maximum and minimum values and the difference between the third quartile and first quartile to summarize variability.

Learn how the variance and standard deviation quantify dispersion by averaging squared distances from the mean, and interpret the standard deviation as the square root in the data's units.

Explore measures of skewness, Fisher, Yule, and Pearson moment coefficients that quantify distribution shape, where zero indicates symmetry and positive or negative values indicate right or left skew.

Explore measures of kurtosis to quantify distribution shape, including Fisher's kurtosis and Pearson's kurtosis, via the fourth moment about the mean and interpretation of pointed versus rounded distributions.

Learn to measure the link between two variables with the correlation coefficient, interpret its -1 to 1 range, and apply it to data like ice creams bought and total spent.

Learn how data visualization transforms raw data into graphical representations, turning mass information into actionable insights and supporting descriptive statistics in the data analysis process.

Explore bar charts and histograms, learn to set axes, plot categories or quantities, and interpret distributions for univariate and bivariate analysis in descriptive statistics.

Explore box plots to visualize distributions using the minimum, first quartile, median, third quartile, and maximum, and compare distributions with real-life examples.

Explore line charts for time series to reveal growth, seasonality, and cross-year comparisons, illustrated with an ice cream shop and stock prices, plus pros and cons.

Explore how stacked area charts reveal the drivers behind revenue trends by splitting values into scoops and using overlay and normalization to uncover cannibalization.

Learn how pie charts convert category shares into angles in a 360-degree circle to show proportions, unlike bar charts of absolute values, with pros and cons illustrated.

Explore how scatter plots visualize correlation between two variables, interpret regression lines and correlation coefficients, and distinguish linear versus non-linear relationships, including pros, cons and example scenarios.

Explore Sankey diagrams to analyze flows and subdivisions of categories, using real-world journeys from login to feed, inbox, and chat to reveal paths that drive purchases and retention.

Explore how a correlation matrix visualizes relationships between variable pairs with color-coded cells, using blue, red, and white to indicate correlation strength, and note that correlation does not imply causality.

Explore a range of data visualizations, from bar charts and histograms to box plots, violin diagrams, heatmaps, 3D graphs, and network graphs, revealing distributions and relationships in data.

Explore data analysis traps and cognitive biases, recognize confirmation bias, and defend against misleading statistics using descriptive statistics, data visualization, and rigorous checks for actionable insights.

Learn why correlation does not imply causation, illustrated by ice cream and sunglasses sales, the sun as a confounding factor, and the need for deeper analysis to infer causality direction.

Explore the dangers of the mean when extreme values skew data, compare with the median, and learn to split populations into cohorts for accurate central values.

Identify the trap of seasonality by zooming out to multi-year data and using line graphs to compare the same months across years; this prevents misinterpreting seasonal trends in sales.

Explore how misleading axes distort line-chart interpretation, using an ice cream shop dataset; compare zero-based versus non-zero axes and add mean or trend lines for clearer insights.

Learn how small sample sizes mislead data analysis and fix it with more data, the law of large numbers, margin of error, confidence intervals, and p values.

Identify selection bias by examining how data samples from loyalty-card customers and online ratings misrepresent the whole population, and learn to limit conclusions to the analyzed subpopulation.

Explore how one dimensional analysis can mislead data insights, and learn to use multidimensional analysis, cross filtering, color coding, and correlation matrices to reveal confounding factors.

Learn the Simpson's paradox, where one dimensional analysis misleads due to a confounding factor like favorite ice cream, revealing positive group correlations but negative overall trends.

Explore how data analysts convert raw big data into actionable business insights by navigating data warehouses, data lakes, and pipelines, while collaborating with data engineers, data scientists, and managers.

Compare spreadsheets, BI tools, analytics platforms, and coding with Python, SQL, and R to find reusable solutions and ask the right questions for data analysis.

Explore how a/b testing optimizes business performance by comparing two variants with random assignment and one variable at a time, while weighing spillover and sample size issues.

Learn how qualitative studies reveal customer feedback beyond data tables by asking broad questions, using questionnaires, mail, and face-to-face conversations, and gathering social media insights to improve a business.

Explore whether you need math, statistics, coding, or artificial intelligence to be a data analyst, while stressing business understanding and asking the right questions.