Students collect data for two or three options of each product (Step Two) and record price, package size, and units, which creates multiple measurements to compare. Students compute unit prices for each option (Day 2, Activity 3) and are instructed to convert units so each ratio represents the same unit of measure, enabling comparison across variable data. Students draw conclusions (Step Four) by choosing an option and explaining choices using the data, and they consider other varying factors (e.g., spoilage, quality) when justifying their decision.

The lesson explicitly defines a statistical question: "A statistical question is designed to collect data. It asks about a specific attribute and expects to have variability in the answers." Students are given multiple activities in which they read questions and circle whether each is statistical or non-statistical and must explain why. Students also create their own statistical questions (e.g., the popcorn example) and identify the attribute and expected variability, and activity pages ask students to determine data type and units for answers.

The lesson explicitly defines a statistical question as one that focuses on an attribute and expects variability (Getting Started). Students practice identifying the attribute and population for multiple questions and decide whether each question should use a whole population or a sample (Student Activity Pages and Answer Key). The Wendy scenario asks students to determine that "Do you have cashiers at your store?" is not statistical and to rewrite it as "How many cashiers are employed at your store?", and the Dr. Franklin activity asks students to choose between yes/no and numeric questions (showing which elicit variable answers).

The lesson repeatedly labels classroom situations as arising from a "statistical question," for example when Mrs. Simms asks campers about favorite meals and when the park designer measures dog heights. Students are asked to identify the attribute being investigated and to decide whether data are numerical or categorical, and they create frequency tables and dot plots that display variability in responses. The Parent Plan explicitly lists the skill "Recognize a statistical question as one that anticipates variability..." as a learning goal.

The lesson repeatedly refers to variability in context: it states Mrs. Simms' attendance data "will likely have a great deal of variability" and contrasts numerical vs. categorical data (e.g., Marla's park equipment question cannot be shown on a stem-and-leaf plot). Students practice interpreting quantitative data from real questions (camp attendance, temperatures, test scores, arcade visitors) by identifying attributes, whether the data are numerical or categorical, and by finding minimums, maximums, and counts from plots. Activity prompts and discussion questions require students to decide if data are population vs. sample and to choose appropriate representations for variable numerical data.

The Getting Started section explains that asking many people their ages can yield many different numbers and shows grouping ages into intervals to manage that variability. Multiple activities have students interpret histograms for variable attributes (camper ages, daily temperatures, employee salaries) and compute frequencies, differences, and totals from grouped data. The Unit 8 Quiz asks students to generate a useful statistical question for deciding when to hold an online meeting, prompting students to formulate questions that consider variability.

Students are asked to work with real-world data sets (weekly allowances, daily flights, swim team ages, movies watched) and to describe distributions by center, spread, and shape. The Parent Plan and Skills statements explicitly state that a data set collected to answer a statistical question has a distribution and that students should relate measures of center and variability to distribution shape. Activities require students to compute mean, median, and mode and to note the effect of outliers on those summaries, which addresses variability in the data.

The Parent Plan Skills section explicitly states that students will "Understand that a set of data collected to answer a statistical question has a distribution which can be described by its center, spread, and overall shape." The lesson repeatedly has students calculate and interpret measures of variability (range, IQR, MAD) and construct/interpret box plots, including prompts such as "What does it mean if a data set has variability?" and activities comparing two orchards and asking which set shows more variability.

Students are prompted to write and use statistical questions such as "How many candies of each color are in a package?" and "What is the average word length in the story?" The activities require students to identify the population and a sample, collect sample data, and use sample results to make inferences about the population. Multiple tasks have students compare different samples (e.g., every 20th vs every 15th word) and observe sampling variability, and the parent/skills sections explicitly list understanding that data collected to answer a statistical question has a distribution and that random sampling supports valid inferences.

The Getting Started section lists multiple two-population questions (e.g., heights of sixth vs. seventh graders; gas mileage of two car types) that model questions asking about variability across a group. Students are prompted to examine graphs and note overlap and shape of distributions, and Activity 2 asks students to compute means and mean absolute deviations and to compare means relative to variability. Several prompts ask students to notice overlap in data values and to consider how sampling methods and variability affect comparisons between populations.

Students are asked explicitly to distinguish statistical from non-statistical questions in multiple activities (e.g., the fishing rodeo item that asks students to circle which question is statistical, Samuel's pool study asking for a non-statistical question, and the Unit Test item asking for a statistical question the manager might ask). The unit overview and Parent Plan list 'How to design a statistical question' and explicitly list the skill 'Recognize a statistical question as one that anticipates variability...' as a learning objective. The answer keys show correct identification of statistical versus non-statistical questions, and students also create and interpret distributions (frequency tables, dot plots) in other tasks.

Students are asked to create a statistical question that involves numerical responses and to record the attribute and population, with guidance on appropriate population size (10–20) and consistent sampling methods. Students collect numerical data, organize it into dot plots/histograms and frequency displays, and compute measures of center and variability (mean, median, mode, range, IQR, mean absolute deviation) as required in Step 4. Students analyze shape and outliers and answer prompts about which measure best represents the data and what the measures of spread tell them, and they must include analysis and inferences that reference the variability in their presentation.

Students gather and analyze numeric data for three distinct populations (Arctic Tundra, Atacama Desert, Himalayan Mountains) across multiple measures (temperature, sunlight hours, wind speed, energy produced, supply weights, costs). They compute and compare energy needs, energy production, fuel-cell shortfalls, and total supply costs for each location and are prompted to use comparative questions (e.g., "Which site had the lowest fuel costs? Which had the most sunlight?") as pieces of math-based evidence for a final recommendation. Tables and tasks require students to produce per-location summaries (per day, per year) and to compare those summaries across locations.

Students run repeated spinner trials (10, 50, 100) and record tallies, calculate experimental probabilities as fractions/decimals/percents, and use those probabilities to predict outcomes for 600 spins. The mixed review and discussion prompts ask why results may change if the experiment is repeated and emphasize how results stabilize with more trials. Vocabulary and short-answer items ask students to explain differences between theoretical and experimental probability and to compute experimental probability from observed counts.

Students make predictions and then collect repeated data (rolling a die 30, 60, 120 times) and compute relative frequencies to compare experimental outcomes with theoretical probabilities. Activities prompt students to use the Law of Large Numbers, to explain discrepancies (e.g., biased equipment, recording error, small sample size), and to build probability models from observed frequencies (non-uniform spinner, fine arts choices). Several tasks ask students to compare what should happen to what actually happens and to justify why results vary.

Students work with a dedicated "Making Inferences" activity that defines a sample and population and asks them to compute proportions from a sample and apply those proportions to a larger population (e.g., 14 of 60 → estimate for 1,143). The lesson explains random sampling and walks students through converting sample fractions to percents and multiplying by the population to produce an estimated count. Several student tasks explicitly ask "What does this data tell me?" and require students to use sample results to predict outcomes for larger groups.

Students formulate hypotheses and run repeated trials in multiple activities (e.g., recording "Pulls Until Blue," "Number of Songs Listened," and trials for the library hunt), collecting data across 10–20 trials. Students compute averages, convert counts to fractions/percents (e.g., count how many trials took 4 or more rolls and write #/20 and percent), and answer reflection questions about variation in their results and reasons for long or short trials. Students use simulations to model real-world questions that inherently produce different outcomes each trial (playlist, visitors, book picks) and explicitly note that results vary because of random chance.

Students generate and evaluate questions about likelihood (e.g., "What are the chances they meet a black dog? A small one? One that's spotted?") and place seven events on a probability line from Impossible (0) to Certain (1). Students build a probability model by computing percentages for each size/color combination from the shelter data (e.g., calculating 9/60 = 15% for medium white) and use those probabilities to answer questions about how likely outcomes are. Students run simulations with a 10-sided die to model how often certain dogs arrive and compare simulated frequencies to model predictions, practicing anticipation and accounting for variability in outcomes.

The Parent Plan lists the goal: "Understand that a set of data collected to answer a statistical question has a distribution which can be described by its center, spread, and overall shape." Students work with many real-world prompts (test scores, commute times, book pages, screen time, push-up/jump-rope counts) and calculate measures of center and variability (mean, median, mode, range, IQR, MAD) and construct/interpret box plots that explicitly account for spread and outliers.

Students work with many data sets that show variability—identifying clusters, outliers, and high vs. low variability (e.g., Tracker A vs. Tracker B, Phone Use vs. Sleep). Students draw and use best‑fit lines to make predictions and are explicitly asked to be cautious when extending predictions beyond the data, which requires attending to variability. Activities ask students to identify relationships across groups (hours studied vs. test scores, age vs. number of toys) and to interpret patterns that arise from many data points.

Students plot paired numerical data many times (hours studied vs. grade, height vs. arm span, temperature vs. ice creams sold, etc.), choose axes/scale/labels, and answer questions about patterns, clusters, and outliers. Students are asked to interpret trends, make predictions from plotted data (e.g., predict grade for 8 hours studied, predict arm span for 74 in), and match real‑world scenarios to scatterplots, which requires thinking about how data vary. Many pages also ask whether the data suggest causation and prompt students to explain variability seen in graphs.

Students are asked in the Bird Migration activity to collect counts of birds across several days, plot each day's data point on a scatterplot, draw a best‑fit line, and form a linear model from the multiple observations. The Parent Note and activity guidance explicitly state that bird counts will naturally vary depending on location, weather, and time spent watching, and encourage discussion about factors that might affect the numbers. Multiple problem pages (e.g., plant growth, ice cream sales) require identifying independent and dependent variables from repeated measurements and using a model to make predictions based on changing inputs.

Students collect survey responses from multiple people (Activity 3 asks students to turn a list of 20 friends' answers into a two‑way frequency table). Students compute and compare proportions using relative frequency tables (Activity 4 and Day 3/Activity 5 show students converting frequency counts to relative frequencies and using those proportions to compare groups). Students design surveys and propose categorical variables (Activity 6 asks students to create a study about music participation and math enjoyment and to design survey questions that produce two‑way tables).

Students are prompted to write two research questions—one numerical and one categorical—and given examples that ask about groups (e.g., how jumping jacks a person can do is related to heart rate; favorite activity vs preferred exercise time). Students are instructed to collect data from multiple people (ask 20 people) and to "try to collect a variety of data," record each person in a row, and use scatterplots and two-way tables to display variability. Reflection prompts ask students to identify outliers and describe patterns and surprises, which has them examine variation in their collected data.

Students compute experimental probability and are asked in question 43b to explain how experimental results may differ from theoretical probability, which prompts consideration of variability in outcomes. Students list sample spaces (question 41) and create tree diagrams (question 46) that require them to enumerate possible varying outcomes. Problems asking how a spinner could model a 40% probability (question 49) and tasks comparing experimental and theoretical probabilities give students occasions to reason about differences across trials.