Difficulty: Introductory | Prerequisites: None (first-year statistics)
Before you can do anything useful in statistics, you need to know what kind of data you are looking at and whether a number describes the whole population or just a sample from it. This topic is the vocabulary layer: every later concept (probability distributions, hypothesis tests, confidence intervals) assumes you can already classify a variable and tell a parameter from a statistic. If these labels are fuzzy, exam questions that look straightforward will trip you up.
Data splits into qualitative (labels, categories) and quantitative (numbers), and quantitative data is either discrete (counted) or continuous (measured). A parameter describes an entire population; a statistic describes a sample drawn from it. Rankings are ordinal, names are nominal.
Parameter
A numerical value that describes a characteristic of an entire population. Think of it as the "true answer" you would get if you could measure every single member of the group.
Statistic
A numerical value computed from a sample (a subset of the population). In simple terms, it is your best estimate of the parameter when you cannot survey everyone.
Qualitative data (categorical data)
Data that describe attributes or categories rather than quantities. The values are labels, not numbers you can add or average. Examples: colour of a chemical sample, brand of phone, political party.
Quantitative data (numerical data)
Data that represent measurable quantities and support arithmetic operations. In simple terms, these are numbers that mean something as numbers.
Discrete data
Quantitative data that result from counting and can only take specific, separated values (typically whole numbers). Think of it as "you can list the possible values and there are gaps between them." Example: the number of crew members on a boat.
Continuous data
Quantitative data that can take any value within a given interval, arising from measurement rather than counting. In simple terms, between any two values there is always another possible value. Example: the mass of a chemical sample in grams.
Nominal data
Categorical data where the categories have no natural order. Names, colours and identification codes are all nominal. License plate numbers look numerical but carry no meaningful magnitude, so they are nominal.
Ordinal data
Categorical data where the categories have a meaningful order, but the distances between ranks are not necessarily equal. A team ranking (1st, 2nd, 3rd) is ordinal: the order matters, but the gap between 1st and 2nd need not equal the gap between 2nd and 3rd.
Qualitative (categorical): describes qualities or categories
Nominal: no natural order (colour, name, ID code)
Ordinal: ordered categories, unequal gaps (rankings, letter grades, satisfaction scales)
Quantitative (numerical): describes measurable amounts
Discrete: counted values with gaps between them (number of students, number of defective items)
Continuous: measured values that can fall anywhere in an interval (weight, temperature, time)
The key test: can you meaningfully add or average the values? Yes = quantitative. No = qualitative. Then: could there be a value between any two observations? Yes = continuous. No = discrete.
A parameter is fixed (though usually unknown). It belongs to the population.
A statistic is calculated from sample data and varies from sample to sample.
Greek letters (mu, sigma) typically denote parameters; Latin letters (x-bar, s) denote statistics.
License plate numbers: look numerical, but they are identifiers with no meaningful magnitude. They are qualitative, nominal.
"Number of objects that are blue": the word "number" makes this quantitative and discrete (it is a count), not qualitative.
Rankings: ordinal, not continuous. The number assigned (1st, 5th) gives order but does not support meaningful subtraction.
Survey percentages from an entire population of interest (e.g. "45% of all TVs in a store cost less than $600"): this describes the whole group, so it is a parameter, not a statistic.
Students often think any number is automatically quantitative data. It is not. Numbers used as labels (license plates, student IDs, phone numbers) are qualitative because arithmetic on them is meaningless.
Students confuse ordinal with continuous because rankings use numbers. The numbers in a ranking only show order; the intervals between them carry no fixed meaning.
Students mix up parameter and statistic by focusing on the type of calculation rather than the source. The same formula (a mean, a proportion) can produce either one. What matters is whether it was computed from the entire population (parameter) or from a sample (statistic).
Expect at least two or three multiple-choice questions that ask you to classify a data set or identify whether a value is a parameter or a statistic. These are easy marks if you have the taxonomy down.
The parameter-vs-statistic distinction also underpins every later topic on sampling distributions, confidence intervals and hypothesis testing. If you cannot tell the two apart here, those sections become much harder.
Ordinal vs nominal is a recurring distractor in multiple-choice options. The exam will offer "nominal" alongside "ordinal" for any ranking question.
True or false: a statistic is a numerical summary of a sample. (True)
True or false: continuous data can take on any value within a given interval. (True)
Fill in the blank: the average income of ALL households in a city is a ____. (Parameter)
True or false: license plate numbers are quantitative data. (False, they are qualitative/nominal)
The number of crew members on a boat is quantitative and ____. (Discrete)
Q: A researcher calculates the average income of all households in New York City. Is this value a parameter or a statistic?
A: It is a parameter. The researcher measured the entire population of interest (all households in the city), not a sample.
Q: Which of the following is quantitative and discrete: (a) mass in grams, (b) license plate numbers, (c) number of crew members, (d) colour of a sample?
A: (c) Number of crew members. It is numerical (quantitative) and results from counting (discrete). Mass is quantitative but continuous. License plates and colour are qualitative.
Q: In a survey, 45% of televisions at an electronics store cost less than $600. Is 45% a parameter or a statistic?
A: A parameter. "All televisions at the store" is the entire population of interest; the 45% describes that whole group.
Q: The ranking of a basketball team is an example of which data type?
A: Ordinal. The ranking carries a meaningful order (1st is better than 5th), but the numerical gap between consecutive ranks is not necessarily uniform.
Q: A data set records the name of each chemical sample. Is this qualitative or quantitative?
A: Qualitative. Names are labels that categorise or identify; they carry no numerical magnitude.
This material connects directly to sampling and experimental design: you need to know that a statistic estimates a parameter before sampling methods make sense. It also feeds into probability distributions, where you will model discrete random variables (counted outcomes) differently from continuous ones (measured outcomes). Every hypothesis test later in the course asks whether a statistic is close enough to a hypothesised parameter value, so the distinction lives through the entire syllabus.
Related Terms / Search Tags
qualitative data, quantitative data, categorical data, numerical data, discrete vs continuous, nominal vs ordinal, parameter vs statistic, population vs sample, levels of measurement, data classification, interval data, ratio data, NYU analytical statistics midterm