Population, Sample, Variables and Data
Population is defined twice, sample three times, variables under two incompatible systems, and data seven times. This page sets out every definition the week gives, marks where they collide, and shows how to write the four terms into a design that will survive examination.
Population: the same word, defined twice in five slides
The key-terms slide defines a population as the "Total category of subjects that is the focus of attention in a particular research project (can be non-human)". Four slides later it is defined again: "the group of people (or items) about which we want to obtain information". The first is category-based and admits non-human subjects; the second is people-first with items in brackets. Neither is marked as a restatement and the two are not word-substitutable.
The same slide adds a requirement — "You need to define your population clearly" — and no criterion of clarity. What it supplies instead is the best-constructed illustration in the week.
The same move decides your study's scope. Projects run by your division and projects delivered into it are two populations, not one thing described twice, and the overlap between them is where scoping arguments in a proposal are lost. The source demonstrates the point once and never generalises from it, so the test below is this page's, not the material's.
Sample: three definitions, three vocabularies
A sample is "A number of subjects drawn from the population" on the key-terms slide; "a group of objects taken from the population for examination" on the sampling slide; and, four sentences later on that slide, "a few of them" — where "them" is the objects you have the resources to collect data on. The three are compatible in intent, incompatible in wording, and none is marked as a restatement.
WHAT THE SOURCE CALLS THE THING YOU COLLECT DATA FROM
| Word used | Where it appears | What it implies |
|---|---|---|
| Subjects | Key-terms slide, in both the population and the sample definitions | The conventional research word; carries no restriction to people |
| People | The collecting-data slide's definition of population | Human by default, with items admitted in brackets |
| Items | The same slide, in brackets after "people" | The non-human residual |
| Objects | The sampling slide, three times in consecutive sentences | Used of an organisation as well as of a thing |
| Members | The sampling slide, and the definition of a census | The unit as a member of a set |
| Respondents | The sampling-bias slide | The first signal in the week that the data come from people who answer questions |
Six words, counted here from the slides; the source states no count and reconciles none of them. Its only acknowledgement that the variation is deliberate is the parenthesis "(can be non-human)", which does not say which word is the general one.
This is not pedantry for a reader in project delivery, because the same slide switches the unit from a person to an organisation without a signal: "You may be interested in obtaining data on an organisation which would be the object of interest in this case." Nothing says whether the organisation is the unit or the source of the data, or whether a sample of organisations is drawn as a sample of people is.
Variables: two systems in one week, neither aware of the other
The slides define a variable as "Characteristics of an item or individual; what you analyse when you use a statistical method". The study notes, in a section reproduced verbatim from a sport-science statistics resource, build something larger: a seven-class typology with examples. That typology is not the teaching material's own writing, and the distinction matters when you cite it.
THE VARIABLE TYPOLOGY IN THE STUDY NOTES, AS GIVEN
| Class | Definition as given | Example given |
|---|---|---|
| Variable | "a set of numbers representing the same kind of thing" — so called "because the numbers vary from subject to subject" | body weight |
| Constant | "If the numbers are the same, the thing is called a constant" | None given, and the term is never used again |
| Numeric | "Variables with numbers as values" | height, weight |
| Nominal | "variables with names or labels as values" | sex, with values male and female |
| Continuous | "can have just about any value to as many decimal places as we like" | height, weight |
| Count | "a variable that is not continuous" | the number of injuries a person has experienced |
| Ordinal | a variable that "can't decide whether it's numeric or nominal", with "an obvious order in the levels" | competitive level: novice, club, national, international |
Reproduced from the study notes, which reproduce it from a sport-science web journal's statistics pages. The recode 1 = novice, 2 = club, 3 = national, 4 = international is a source example with four levels; it is not a scale or a recommended coding.
Set this beside the same week's measurement slide, which lists scales of measurement — nominal, ordinal and ratio — and the two systems share two words and disagree about the rest. The notes have no ratio and no interval; the slides have no continuous, no count and no constant. The two also define "ordinal" differently: the notes make it a variable whose type is indeterminate, the slides a matter of the values being ranked. That conflict is worked through at Measurement and the Basis of Validity.
Data: seven definitions in one week
The word is defined seven times across the week's two documents and none of the seven is reconciled with another. They are set out below because the count is the finding: this is not a term the material is careless with in one place, it is a term it has no settled meaning for.
EVERY DEFINITION OF DATA THE WEEK SUPPLIES
| # | Where | Definition as given |
|---|---|---|
| 1 | Key-terms slide | "The different values associated with the variable" |
| 2 | Managing your data slide | "Data is the evidence the researcher gathers for the research project." |
| 3 | Study notes, borrowed statistics section | "Data are usually just a set of numbers." |
| 4 | Study notes, a university research-data policy | "data in the form of facts, observations, images, computer program results, recordings, measurements or experiences on which an argument, theory, test or hypothesis, or another research output is based" |
| 5 | Study notes, a second university's draft policy | "Data are facts, observations or experiences on which an argument, theory or test is based." |
| 6 | Study notes, a third university's policy | "data, records, files or other evidence, irrespective of their content or form … including primary materials and analysed data" |
| 7 | Study notes, a fourth university's schedule | "factual records … used as primary sources for research, and that are commonly accepted in the research community as necessary to validate research findings" |
Definitions 4 to 7 are institutional policies the notes quote from a national research-data service guide; the institutions are scrubbed. They are presented as agreeing, without comparison, and only the last supplies a test of inclusion.
The first is variable-bound and formal; the second is evidentiary and project-bound. Values associated with a variable and evidence gathered for a project are not the same class of thing. Where this page needs a working sense of the word it uses the first, because the variable typology depends on it. That is a working choice and not a ruling — it is one of the seven, not the correct one.
The worked data set, and the two defects in it
The study notes supply one worked data set and build the rest of the section on it. It is introduced as "a data set with three nominal variables, two continuous numeric variables, one ordinal variable, and one counting variable". The stated count matches the table exactly — seven columns, seven accounted for — which is a rarer event in this material than it should be.
THE WORKED DATA SET, REPRODUCED AS SUPPLIED
| subject | weight | height | sex | sport | level | injuries |
|---|---|---|---|---|---|---|
| AJH | 63 | 170 | female | swimming | 3 | 0 |
| CJD | 78 | 185 | male | basketball | 2 | 0 |
| NIH | 68 | (blank) | female | basketball | 5 | 2 |
| ERF | 69 | 177 | male | (blank) | 1 | 1 |
| MBA | etc. | etc. | etc. | etc. | etc. | etc. |
Every value is a source example — not a finding, not a norm, not a typical value. The two blanks are deliberate and demonstrate missing values. The three-letter subject identifiers are never explained, in a week that elsewhere requires data to be protected; nothing comments on their identifiability.
The second defect sits a few paragraphs later, in the passage on summarising a nominal variable. The text reads: "a group of athletes might consist of 101 basket-ballers, 49 footballers, and 51 others", and says the frequencies can be shown "as proportions in a pie chart, as shown". Those counts are source examples in an imagined data set — not sample sizes, thresholds or benchmarks — and they sum to 201.
A third, smaller failure sits in the same paragraph: the text calls "the new sport football" an outlier in need of correction, when football is already one of its three categories with 49 members. The point survives only if the stray value is a misspelling of an existing category, and no misspelling is printed.
Writing the four terms into your design
The sequence below is this page's construction. The source supplies the terms and one worked population definition; it supplies no procedure for getting from a research question to a defined population, sample and variable list, and none of these steps should be attributed to it.
From research question to a defensible set of four terms
Name the unit of analysis in one noun
Project, team, contract, manager, organisation. Write it once and use the same word throughout. The source uses six; you get one.
Write the population as that noun plus every qualifier
Apply the boundary test above. Name a case that satisfies the noun but fails a qualifier, and a case that satisfies the qualifiers but is not the noun.
State whether you are studying the population or a sample of it
If everything, you are running a census, which the week defines and never returns to. If not, see Whole Population or Sample?.
List your variables and give each one a class
Use one classification scheme and name it. Mark which variables you will read off records and which are answers people will give you.
Say what will count as data, in your own sentence
One sentence: what will be recorded, in what form, about which unit. Do not leave a reader to pick from seven meanings.
Distinguish the parameters you want from the statistics you will get
A measure describing the population is a parameter; the same measure on your sample is a statistic. The week defines both and never performs the move from one to the other.
What to carry forward
- Population and sample are each defined more than once here, in vocabulary that does not substitute. Quote one and note the other.
- Pick one word for your unit of analysis and hold it. The source uses six.
- Two variable classification systems sit in one week without a crosswalk. Name the one you used.
- Seven definitions of data appear in the week. Using one is a working choice, not a finding that it is correct.
- The worked data set carries an out-of-range level value and a chart that contradicts its own counts. Report both; repair neither.
Frequently asked questions
Which definition of population should I use in my methodology chapter?
Either of the two the week gives, provided you quote the one you used and note that the material supplies a second in different terms. The category-based definition explicitly admits non-human subjects, which is usually the more useful one for project research where the unit is a project or a contract rather than a person.
What does the source say a variable is?
The slides say a variable is a characteristic of an item or individual, and what you analyse when you use a statistical method. The study notes, quoting a sport-science statistics resource, define it as a set of values representing the same kind of thing, so called because the values vary from subject to subject. The two are compatible; the classification systems built on them are not.
Why does this page list seven definitions of data instead of giving me one?
Because the week supplies seven and reconciles none of them, and choosing between them here would attribute a decision to the source that the source never made. Where a working sense is needed, this page uses the variable-bound definition and says so; that is a working choice, not a ruling about which of the seven is right.
Is the worked data set safe to reuse as a teaching example?
Only with its defects declared. It contains a competitive level of 5 when its own text defines four levels, and the pie chart built from the same section shows a largest segment that contradicts the counts printed beside it. The stated variable count does match the table, which is worth noting because it is unusual in this material.
Does the week explain how to sample organisations rather than people?
No. It raises the organisation as a possible object of interest in a single sentence and never relates any of its six sampling types to organisational units. Decide and state your unit of analysis yourself, and take the sampling procedure from a source that addresses the unit you have chosen.
References and source attribution
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- Veal, A. J. 2005, Business Research Methods: A Managerial Approach, Longman.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Hopkins 2000, 'Quantitative Research Design', a sport-science web journal, vol. 4, issue 1 — journal title, address and author given names scrubbed. Set as this week's required reading; the same author's statistics pages are the source of the variable typology and the worked data set reproduced in the study notes.
- The supplied teaching source: the slide deck on the nature of quantitative research and both copies of the accompanying study notes, including the four institutional research-data policy definitions the notes quote from a national research-data service guide.
Suggested questions for Ask KEVOS
- Draft the population and sample paragraphs of my methodology chapter and show me where the boundary is ambiguous.
- Classify every variable in my data collection instrument and tell me which classification system I am using.
- Write one sentence defining what will count as data in my project, for use in my ethics application.
- My unit of analysis is a project rather than a person. What does this material fail to tell me?
- Show me the difference between a parameter and a statistic using figures from my own study.