Chapter 4 — Presentation Of Data

What is presentation of data? Presentation of data means arranging collected facts and figures in a systematic, meaningful form so that underlying patterns, trends and relationships become clear and easy to interpret. It converts raw numerical or categorical information into tables, diagrams and graphs.

Why is it important?

• Simplifies complex data and saves time.
• Makes comparison and interpretation easier.
• Helps reveal trends, central tendencies and variability.
• Aids decision-making and communication of findings.

Types of data relevant to presentation

• Qualitative (categorical): e.g., occupation, region, sector. Best shown by bar diagrams and pie charts.
• Quantitative:
  • Discrete (countable): e.g., number of children in households. Use frequency tables, bar charts.
  • Continuous (measurable): e.g., height, marks, income. Use frequency distribution, histogram, ogive, frequency polygon.
• Primary vs Secondary: Primary data are collected firsthand; secondary are from published sources. Presentation methods are same for both.

Basic steps in presentation

1. Classify data (group into classes if continuous or numerous discrete values).
2. Construct a frequency distribution (absolute frequencies, relative frequencies, cumulative frequencies).
3. Choose appropriate diagram/graph depending on the type of data and purpose.

Key ideas to remember when presenting data

• Label axes, give a clear title, units of measurement and a legend if needed.
• Choose suitable class-widths; avoid too many or too few classes.
• For unequal class widths in a histogram use frequency density (area proportional to frequency).
• Avoid misleading visuals (inaccurate scales, truncated axes, unnecessary 3D effects).

How presentation helps analysis — using tables and graphs you can quickly identify central tendencies (mean/median/mode), spread (range, quartiles) and patterns such as seasonality, growth/decline and concentration.

Definition: Data are facts, numbers or information collected for reference, analysis and decision making. In economics, data form the basis for statistical analysis and presentation.

Main types of data:

• By source
  • Primary data: Collected first-hand for a specific purpose (e.g., survey responses, experiments).
  • Secondary data: Collected earlier by someone else (e.g., census reports, published articles).
• By nature (qualitative/quantitative)
  • Qualitative (categorical): Non-numeric attributes (e.g., gender, religion, occupation). Subtypes: nominal (no order) and ordinal (with order).
  • Quantitative (numerical): Numeric measurements. Subtypes: discrete (countable integers) and continuous (measurable on a continuum).
• By time
  • Cross-sectional: Data collected at a single point of time across units (e.g., household consumption in 2023 across states).
  • Time-series: Observations on the same variable at successive time points (e.g., GDP each year from 2000–2020).
• By form
  • Ungrouped (raw) data: Each observation listed separately (e.g., marks of 30 students).
  • Grouped data: Observations organized into class intervals or categories with frequencies (e.g., heights grouped into 150–155 cm, 155–160 cm, ...).

Principles / Rules for classification

• Classes or categories must be mutually exclusive (no overlap).
• They must be collectively exhaustive (cover all observations).
• Classes should be simple and homogeneous and suitable for the purpose of analysis.
• Class intervals are usually of equal width (unless justified otherwise).

Important derived concepts

• Frequency (f): number of observations in a class or category.
• Relative frequency: fraction/ proportion of total: f/N.
• Cumulative frequency: running total of frequencies up to a class.

Why classify? Classification groups data to reveal patterns, simplify analysis and make graphical representation meaningful (e.g., making histograms, ogives, pie charts).

What is Tabular Presentation?

Tabular presentation is a method of arranging raw data in rows and columns (a table) so that it becomes organised, easy to read and ready for analysis. A well-constructed table summarises data, highlights patterns and supports comparison.

Components of a table

• Title: concise and descriptive.
• Headings/Column titles: name of variables and units of measurement.
• Stub (row labels): labels for rows (e.g., categories or class intervals).
• Body: cells containing the data or frequencies.
• Footnote/Source: any clarifications or data source.

Types of tables

• Simple table: Records individual observations (ungrouped data).
• Frequency distribution table: Groups observations into classes and lists frequencies (grouped data).
• Two-way (contingency) table: Shows joint distribution of two variables (e.g., gender × pass/fail).

Steps / Rules to construct a good table

• Give a clear title and state the period/units.
• Arrange categories logically (chronological, numeric, or alphabetical order).
• Use appropriate class intervals (mutually exclusive, exhaustive, preferably equal width for histograms).
• Include totals (e.g., total frequency).
• Keep layout neat: align numbers (usually right-aligned) and use consistent precision (decimal places).

Advantages and limitations

• Advantages: simplifies large data sets, reveals patterns, makes comparison easier, is a basis for graphical presentation.
• Limitations: grouping causes loss of individual detail; poor choice of classes can hide information.

Small example table (marks of 10 students)

Example: Frequency distribution (grouped)

Such tables are the starting point for graphs like histograms, frequency polygons and ogives.

What are Diagrammatic Methods?
Diagrammatic methods are graphical techniques used to present statistical data visually so patterns, trends and comparisons become easier to understand. They convert numerical tables into charts or diagrams (bars, pie, histograms, polygons, ogives, line graphs, pictograms) that communicate information quickly.

Why use them?

• Make large data sets easier to interpret.
• Highlight trends, comparisons and composition at a glance.
• Useful for report writing, presentations and decision making.

General rules for good diagrams

• Choose the diagram that suits the data and purpose (comparison, composition, distribution or time series).
• Label axes, units and categories clearly; give a title and source.
• Use uniform scale; avoid 3D effects that distort perception.
• Provide a legend if there are multiple series or symbols.
• If using pictograms, state the value represented by one symbol.

Common diagram types and how to prepare them

1. Bar Diagram — used for comparing discrete categories (e.g., sales by state)

• Axes: categories on x-axis, frequency/value on y-axis.
• Bars should be of equal width and equidistant (vertical or horizontal).
• Variations: simple, multiple (grouped) bars for comparing sub-groups, and component (stacked) bars to show parts of a whole.

2. Pie Diagram (Pie Chart) — shows composition (parts of a whole)

• Convert each category to an angle: Angle = (Category value / Total) × 360°.
• Draw slices proportionally; include labels or a legend with percentages.

3. Histogram — for continuous data grouped into class intervals (e.g., income, age)

• Plot class intervals on x-axis and frequency on y-axis as adjacent rectangles (no gaps).
• If class widths are unequal, use frequency density = frequency / class width for heights.

4. Frequency Polygon — alternative to histogram to show distribution shape

• Use class mid-points on x-axis and plot frequencies as points; join points with straight lines.
• Extend polygon to the x-axis at ends (by adding a class with zero frequency) to close the shape.

5. Ogive (Cumulative Frequency Curve) — shows cumulative frequencies and is useful for median/percentile estimation

• Plot cumulative frequency against upper class boundaries (or lower boundaries) and join points with a smooth curve or straight lines.

6. Line Graph (Time Series) — shows trends over time (months, years)

• Plot time on x-axis and variable on y-axis; mark data points and join with straight lines.

7. Pictogram — uses pictures/icons to represent quantities (good for simple public displays)

• One symbol must represent a fixed quantity; include a key (e.g., 1 icon = 100 students).

Advantages

• Quick visual impression, easier comparison and interpretation.
• Can reveal patterns (trend, seasonal effects, skewness) not obvious in tables.

Limitations

• Poorly drawn diagrams can mislead (incorrect scales, truncated axes, 3D effects).
• Not suitable for very detailed numerical analysis; best for summary presentation.

Practical steps when preparing a diagram

1. Decide the objective (compare categories, show composition, show distribution or trend).
2. Choose the appropriate diagram type.
3. Compute necessary derived values (percentages, angles, midpoints, frequency densities, cumulative frequencies).
4. Draw axes, mark scale carefully and plot; label everything and add a title and source.

Graphical methods are visual techniques used to present statistical data so patterns, trends, comparisons and distributions become easy to understand. They convert numbers into pictures (bars, lines, circles, polygons, etc.), which help in quick interpretation and communication.

Key principles when drawing any graph:

• Choose the appropriate type of graph for the data (categorical, discrete, continuous, time-series, bivariate).
• Label axes with variable names and units, give a clear title, and include a legend if needed.
• Select a suitable scale so the entire data range fits and the graph is not distorted.
• Use class boundaries (continuous scale) for histograms and ogives, and mid-points for frequency polygons.

Main graphical methods:

• Bar Diagram — Simple bars for categorical or discrete data. Bars are separated by gaps. Variants: multiple bar (two or more series side-by-side) and sub-divided/stacked bar (components stacked within a single bar).
• Pie Chart — Circle divided into sectors showing proportional composition. Useful for parts of a whole.
• Histogram — Used for continuous grouped data. Adjacent bars with no gaps. Height represents frequency density when class widths differ.
• Frequency Polygon — Points plotted at class mid-points with ordinate equal to frequency (or frequency density) and joined by straight lines. Useful for comparing distributions.
• Ogive (Cumulative Frequency Curve) — Plots cumulative frequency against class boundaries. Two types: "less than" ogive and "more than" ogive. Intersection gives median and quartiles.
• Line Graph (Time Series) — Points for observations over time joined by lines. Ideal for showing trends.
• Scatter Plot — Plots paired (x,y) data to show relationship or correlation between two variables.
• Pareto Chart — Bars in descending order of frequency and a cumulative percentage line; highlights the most important factors (80/20 rule).

Interpretation tips:

• Shape of distribution (symmetric, positively/negatively skewed) can be seen from histograms and frequency polygons.
• Ogive helps locate median (point where cumulative frequency = N/2) and quartiles (N/4, 3N/4).
• Scatter plots indicate direction and strength of relationship; add a trend line to see correlation.

Overview
"Construction Steps and Conventions" describes how to convert raw data into classed frequency tables and graphical displays (histogram, frequency polygon, ogive, bar chart, pie chart etc.) using standard steps and rules so graphs are accurate, comparable and not misleading.

1. Collect and sort data
   Obtain the raw data and arrange it in ascending order (if individual values) or prepare an initial tally.
2. Decide number of classes (k)
   Use judgment or a rule (e.g. Sturges' rule: k ≈ 1 + 3.322 log10 n). Typical k is between 5 and 15 depending on sample size.
3. Compute range and class width
   Range = max − min. Class width (h) ≈ Range / k. Round h to a convenient number (1, 2, 5, 10 …) so classes are simple.
4. Form class intervals
   Choose a suitable starting point and make k non-overlapping intervals of equal width (unless unequal widths are needed). Write class limits (e.g. 10–19, 20–29).
5. Convert to class boundaries (for continuous presentation)
   For continuous graphs (histograms, ogives) use class boundaries so adjacent classes touch. If original classes were 10–19 and 20–29 (discrete limits), boundaries become 9.5–19.5 and 19.5–29.5 (subtract/add half the measuring unit).
6. Tabulate frequencies
   Count observations in each class → frequency (f). Also compute cumulative frequency (CF), relative frequency (f/N) and percentage frequency.
7. Adjust for unequal class widths
   If class widths differ, use frequency density = frequency / class width on the vertical axis for histograms so area represents frequency.
8. Choose representation and draw
   • Decide histogram, frequency polygon, ogive, bar chart, pie chart, line graph or Pareto chart.
   • Draw axes, choose an appropriate scale (use simple numbers; ensure axis for bar/hist starts at zero), plot points/rectangles/angles, join where needed, add title, labels, legend and source.
9. Conventions to follow
   • Bars touch in histograms (continuous data); bars separated in bar charts (discrete categories).
   • Use class boundaries for histograms and ogives; use class marks (mid‑points) for frequency polygons.
   • Vertical axis normally starts at zero for bar/histogram to avoid misleading visual effects.
   • When class widths unequal, plot frequency density (height) so area ∝ frequency.
   • Label axes clearly (variable and units), give a descriptive title and provide source of data.

How it looks in practice (short workflow): Raw data → choose k and h → create class intervals and boundaries → tally frequencies → compute CF/relative frequencies/frequency density (if needed) → choose chart → draw axes with scale → plot and label.

Definition: Interpretation means explaining what the presented data shows (patterns, direction, comparisons). Analysis goes further: it examines causes, relationships, reliability and implications of the data and suggests conclusions or actions.

Purpose: To convert numbers into meaningful information for decision-making — e.g., identifying trends, comparing groups, detecting outliers, and testing relationships.

Step-by-step approach

• Understand the data: units, period, population/sample, base (for index numbers) and any missing values.
• Summarise appropriately: use totals, averages, percentages, rates, or index numbers depending on the question.
• Visualise: choose a graph that matches the aim (trend, composition, distribution, relationship).
• Compute measures: central tendency (mean/median/mode), dispersion (range, standard deviation, coefficient of variation), growth rates or index numbers as needed.
• Interpret patterns: direction (increasing/decreasing), seasonality, peaks/troughs, anomalies/outliers and relative comparisons (use percentages or per-capita measures for fair comparison).
• Analyse causes and implications: relate observed patterns to possible economic, social or policy factors and assess reliability (sample size, data quality, scale or aggregation bias).
• Conclude with clear, evidence-based statements and, if required, recommendations.

Common pitfalls to watch for

• Misleading scales or truncated axes in graphs.
• Confusing absolute values with relative change — use percent change for comparison over time or across sizes.
• Aggregating heterogeneous groups without standardisation (per capita or rates may be needed).
• Ignoring outliers or small sample sizes when generalising.

How interpretation and analysis are used in economics

• Policymakers read time-series data (GDP, inflation) to decide fiscal or monetary measures.
• Firms analyse sales trends and seasonality to plan production and inventory.
• Researchers use scatter plots and regression to test relationships (e.g., income vs consumption).

What this topic covers: Practical Applications and Exercises in "Presentation of Data" teaches how to convert raw observations into organised tables and graphs that make patterns visible and support interpretation. You learn to form frequency distributions, compute relative and cumulative frequencies, choose suitable graphical forms (bar chart, histogram, frequency polygon, ogive, pie chart, time-series line), and interpret results to draw conclusions.

Step-by-step practical procedure:

1. Collect and inspect raw data: Check for measurement units, discrete vs continuous nature, and any obvious errors/outliers.
2. Decide grouping: For large/raw numeric data, form class intervals. Prefer 5–15 classes depending on sample size. Use equal class width if possible.
3. Compute class boundaries and mid-points: Convert limits to continuous class boundaries (adjust by half the smallest measurement unit, commonly 0.5 when units are whole numbers). Mid-point = (lower limit + upper limit)/2.
4. Build frequency table: Tally frequencies, then compute relative frequencies (f / N), cumulative frequencies, and frequency density if class widths vary (frequency density = f / class width).
5. Choose the right visual form: Qualitative/discrete—bar chart or pie chart. Continuous—histogram (for frequency), frequency polygon (mid-points joined), ogive (cumulative frequency). Time-series—line graph.
6. Draw carefully: Label axes, give a title, choose an appropriate scale, and for histograms use contiguous bars (no gaps). For unequal widths use frequency density on vertical axis.
7. Interpret and conclude: Look for central tendency, spread, skewness, modes, peaks, clusters, gaps, and trends over time. Relate findings to the context (e.g., policy implication, business decision).

Exam & classroom tips:

• Always show class limits, boundaries, mid-points and formula steps when asked. Marks are awarded for method and presentation.
• When class widths are unequal, use frequency density for histograms; otherwise heights will mislead.
• To find median or mode from grouped data, use the standard grouped-data formulae (show each symbol and substitution clearly).
• Use ogive to locate median graphically: the median corresponds to cumulative frequency = N/2 on the vertical axis.

Common practical exercises students should practice: constructing frequency distributions, drawing histograms, frequency polygons and ogives, converting a bar chart to a pie chart (percentages), finding mean/median/mode from grouped frequency tables, and interpreting graphs in context.