Ask a question that expects variation

Try

A school wants to improve its morning timetable. Before collecting numbers, compare “When did Maya arrive today?” with “How do arrival times vary among students during a typical school week?” The first requests one fact. The second anticipates differences across people and days, making it a statistical question. Write your own version that identifies whose experiences matter and which dates count.

Predict

Predict whether recording more timestamps automatically produces a better answer. A thousand measurements of one student may describe that student carefully while saying little about everyone else. Also predict whether a missing arrival time should be recorded as zero minutes. Zero is a possible value with a meaning; missing means the value was not observed.

Experiment

Use this original synthetic recording plan. Each row represents one student on one scheduled day, not a new independent student.

Student	Day	Minutes after opening	Travel mode
A	Monday	12	Bus
A	Tuesday	missing	Bus
B	Monday	4	Walk
B	Tuesday	8	Walk

Count four scheduled records, three observed times and two students. The mean of the observed times is (12+4+8)/3=8 minutes. Replacing the missing reading by zero gives 6, an invented result answering a different question. Add a column specifying why the value is missing, if that reason can be collected responsibly.

Understand

An observational unit is the entity being described: here a student-day. A variable is an attribute recorded for that unit. The population is the full group the question concerns; the sample is the observed subset. Repeated rows from one person may be related. A row count alone cannot establish independent information.

Quantitative variables describe meaningful amounts: elapsed minutes or number of late arrivals. Counts are discrete; a measurement such as duration is commonly modeled as continuous even though a device rounds it. Categorical variables identify groups, including ordered ratings such as low, medium and high. Numeric codes do not automatically create quantitative measurements. Averaging bus route numbers or student identification numbers has no sensible interpretation.

Operational definitions turn an idea into a reproducible measurement. “Arrival” might mean entering the gate, reaching a classroom or registering attendance. Choose a rule before comparing groups. Different rules can create apparent differences unrelated to the phenomenon of interest. Document units, rounding, eligible days and exceptional events.

An explanatory variable is used to describe or predict differences in a response variable. Calling travel mode explanatory does not prove it causes arrival time. Distance, timetable and family circumstances may differ too. A useful investigation proceeds from question to collection plan, analysis and a qualified interpretation, with opportunities to revise the question when the data cannot answer it.

Apply

Draft a five-line data dictionary for your question: unit, eligible population, response definition, explanatory variable and missing-value rule. Then exchange it with someone who has not heard your plan. If they cannot collect comparable observations from the instructions, refine the definitions. The practice tasks ask for this reasoning as well as arithmetic; matching a number alone does not validate a measurement process.

Distinguish a count, a measured quantity and its recorded precision

Changed-case classification. A transport study records (a) the number of late buses in one hour, (b) the exact elapsed journey time modeled before rounding, (c) that journey time rounded to the nearest whole minute, and (d) a route label such as 24. Classify each variable and justify the classification by its possible values, not merely by whether digits appear.

Feedback

(a) is a discrete quantitative count: its possible values are nonnegative integers. (b) is modeled as continuous quantitative: elapsed time can lie anywhere in an interval. (c), the actual rounded recording, takes discrete whole-minute values, even though the underlying quantity in (b) is continuous. (d) is categorical: the digits name a route, and adding or averaging route codes has no useful measurement interpretation. A complete explanation distinguishes the underlying quantity from recording precision and identifies the unit or meaning in context. Numeric appearance alone does not determine the type.

Continue

Next examine how units enter the sample. Keep the distinction between many records and many independently selected people: it becomes central when assessing representativeness and uncertainty.