29  Quantitative and Qualitative Data

The most basic distinction in data analysis is between quantitative (numerical) and qualitative (categorical) data. The distinction determines which descriptive statistics, graphs, and inferential tests are appropriate.

29.1 Quantitative Data

Quantitative data are numerical measurements — values that can be added, averaged, and ordered.

TipTwo Sub-types of Quantitative Data
Sub-type Description Examples
Discrete Countable; takes only specific values (usually integers) Number of children, vehicles registered, cases reported
Continuous Takes any value on a continuum Height, weight, time, temperature, income
TipDistinguishing Discrete from Continuous
  • Discrete: “How many?” — answer is a whole number.
  • Continuous: “How much?” — answer can be a decimal of any precision.

A household has 2 or 3 children, never 2.5 (discrete). A person can be 165.4 cm tall (continuous).

29.2 Qualitative (Categorical) Data

Qualitative data describe categories or attributes — values that name a quality rather than a quantity.

TipTwo Sub-types of Qualitative Data
Sub-type Description Examples
Nominal Categories without a natural order Religion, gender, blood group
Ordinal Categories with a natural order, but unequal gaps Educational attainment (school / graduate / postgraduate); satisfaction (Likert)
TipDistinguishing Nominal from Ordinal
  • Nominal: Asking “is this rank higher than that?” makes no sense — male is not “higher” than female.
  • Ordinal: Ranking is meaningful — postgraduate is “higher” than school in attainment, but the gap from school-to-graduate is not necessarily equal to graduate-to-postgraduate.

29.3 The Four Levels of Measurement Together

Stevens (1946) integrates quantitative and qualitative as four levels.

TipStevens’s Four Levels — NOIR
Level Type Operation that becomes meaningful Example
Nominal Qualitative = and ≠ Gender, religion
Ordinal Qualitative < and > Education level, satisfaction
Interval Quantitative + and − Temperature in °C, year
Ratio Quantitative + − × ÷ Height, weight, income, age
TipThe Crucial Difference between Interval and Ratio
  • Interval scale has equal intervals but no absolute zero. 0°C does not mean “no temperature”.
  • Ratio scale has equal intervals and an absolute zero. 0 kg means “no weight”.

Hence, on a ratio scale we can say “20 kg is twice 10 kg”; on an interval scale, “20°C is not twice 10°C”.

29.4 Choosing Statistics by Data Type

TipAppropriate Statistics for Each Level
Level Central tendency Dispersion Common test
Nominal Mode None typical χ² (chi-square)
Ordinal Median, Mode Range, IQR Mann-Whitney, Kruskal-Wallis
Interval Mean, Median, Mode Range, SD, Variance t-test, ANOVA
Ratio Geometric mean, Mean All measures All parametric tests 

flowchart TB
  D[Data] --> Q[Quantitative]
  D --> C[Qualitative]
  Q --> DC[Discrete]
  Q --> CT[Continuous]
  C --> NM[Nominal]
  C --> OR[Ordinal]
  NM --> NOM[NOIR — Nominal]
  OR --> ORL[NOIR — Ordinal]
  DC --> INT[NOIR — often Interval]
  CT --> RAT[NOIR — Interval / Ratio]
    classDef default fill:#003366,color:#ffffff,stroke:#ffcc00,stroke-width:3px,rx:10px,ry:10px;

29.5 Worked Examples — Classifying Variables

TipWorked Examples — assign each variable a level
  • Number of children in a household → Quantitative · Discrete · Ratio.
  • Height in centimetres → Quantitative · Continuous · Ratio.
  • Gender (M / F / Other) → Qualitative · Nominal.
  • Education level (school / graduate / PG) → Qualitative · Ordinal.
  • Satisfaction on a 5-point Likert scale → Qualitative · Ordinal (often treated as Interval in practice).
  • Year of birth → Quantitative · Discrete · Interval (no absolute zero — year 0 is arbitrary).
  • Income in rupees → Quantitative · Continuous · Ratio.
  • Temperature in Celsius → Quantitative · Continuous · Interval.
  • Religion → Qualitative · Nominal.

29.6 Conversion Between Types

It is often necessary to convert one data type to another for analysis.

TipCommon Conversions
  • Continuous → Ordinal (binning): Income (numeric) → Income brackets (low / middle / high).
  • Ordinal → Nominal: Education levels → “Has graduate degree” (yes / no).
  • Quantitative → Categorical (dichotomisation): Convert age into “≥ 60” vs “< 60”.
  • Categorical → Numeric (dummy coding): Gender → 0 / 1; Religion → multiple 0/1 dummies.

Each conversion loses information. Going from continuous to categorical reduces statistical power; going from categorical to numeric for ordinal scales makes assumptions that may not hold.

29.7 Practice Questions

Q 01 Quantitative vs Qualitative Easy

Which of the following is qualitative data?

  • AIncome in rupees
  • BHeight in centimetres
  • CReligion
  • DNumber of children
View solution
Correct Option: C
Religion is a categorical attribute = qualitative. The others are numerical.
Q 02 Discrete vs Continuous Easy

"Number of cars produced in a factory each day" is which type of quantitative data?

  • ADiscrete
  • BContinuous
  • CNominal
  • DOrdinal
View solution
Correct Option: A
Number of cars takes whole-number values = discrete.
Q 03 Nominal Data Medium

Which is the most appropriate measure of central tendency for nominal data?

  • AMean
  • BMedian
  • CMode
  • DGeometric mean
View solution
Correct Option: C
Only the mode is meaningful for nominal data — you cannot order or average categories.
Q 04 Interval vs Ratio Medium

Why is "temperature in Celsius" considered an *interval* rather than a *ratio* scale?

  • AIt is qualitative
  • BIt does not have an absolute zero point
  • CIt cannot be measured
  • DIt has no equal intervals
View solution
Correct Option: B
0°C does not mean "no temperature"; it is a conventional zero. Hence Celsius is interval, not ratio.
Q 05 Ordinal Data Medium

Which of the following is an example of ordinal data?

  • AHair colour
  • BAge in years
  • CEducation level (school / graduate / postgraduate)
  • DNumber of siblings
View solution
Correct Option: C
Education level has a natural rank order but unequal "intervals" between levels = ordinal.
Q 06 Likert Scale Medium

A 5-point Likert scale (Strongly Agree → Strongly Disagree) is *most strictly* classified as:

  • ANominal
  • BOrdinal
  • CInterval
  • DRatio
View solution
Correct Option: B
Strictly ordinal — though often treated as interval in practice for averaging.
Q 07 Levels Hard

Match each variable with its level of measurement:

(i) Postal code (a) Ratio
(ii) Income in rupees (b) Interval
(iii) Year of birth (c) Nominal
(iv) Marathon-finishing rank (d) Ordinal
  • A(i)-(c), (ii)-(a), (iii)-(b), (iv)-(d)
  • B(i)-(a), (ii)-(b), (iii)-(c), (iv)-(d)
  • C(i)-(b), (ii)-(c), (iii)-(d), (iv)-(a)
  • D(i)-(d), (ii)-(c), (iii)-(b), (iv)-(a)
View solution
Correct Option: A
Postal code → nominal (just labels); Income → ratio (absolute zero); Year of birth → interval (no absolute zero); Rank → ordinal.
Q 08 Conversion Easy

Converting age in years (continuous) into "young / middle / old" categories is called:

  • AImputation
  • BBinning / Categorisation
  • CNormalisation
  • DStandardisation
View solution
Correct Option: B
Binning (or categorisation) groups continuous values into ordinal/categorical bands.
ImportantQuick recall
  • Quantitative: Discrete (counts) vs Continuous (continuum).
  • Qualitative: Nominal (no order) vs Ordinal (ordered).
  • Stevens’s levels: Nominal, Ordinal, Interval, Ratio (NOIR).
  • Interval has equal gaps but no absolute zero; Ratio has both.
  • Statistics by level: nominal → mode, χ²; ordinal → median, non-parametric tests; interval/ratio → mean, t-test, ANOVA.
  • Likert scale is strictly ordinal; often treated as interval.
  • Year is interval (no absolute zero); Age is ratio.