After data have been organised into frequency distributions, the next task is to present them in a clear and attractive manner so that their salient features can be grasped quickly. A well-presented table or diagram communicates in seconds what pages of numbers cannot. Presentation of data makes the data easy to understand, facilitates comparison, and prepares the ground for analysis. Poorly presented data, by contrast, confuse the reader and hide the truth.
Data can be presented in three broad ways: textual presentation, tabular presentation, and diagrammatic or graphical presentation. Textual presentation describes data in words and is suited to small data sets. Tabular presentation arranges data in rows and columns, which is the most common and rigorous method. Diagrammatic and graphical presentation uses bars, pies, lines, histograms, polygons and ogives to make the data visually striking and easy to compare.
This chapter explains the essentials of a good table, the components of a table, the general rules of tabulation, the different kinds of diagrams such as bar diagrams and pie charts, and the graphical methods such as histograms, frequency polygons and ogives, along with their construction and uses.
Textual presentation is the presentation of data in the form of descriptive statements in words. For example: "The production of rice in India was 100 million tonnes in 2020, 105 million tonnes in 2021 and 110 million tonnes in 2022." This method is suitable when the volume of data is small and only a few figures are to be communicated.
The advantages of textual presentation are that it is simple, easy to write and requires no special skill. However, its disadvantages are serious: it is dull, it cannot handle large data sets, it makes comparison difficult, and it consumes much time and space. Hence textual presentation is rarely used alone; it is usually combined with tables and diagrams.
Tabular presentation is the arrangement of data in rows and columns, which is the most common and precise method of presenting statistical data. A statistical table is a systematic organisation of data in which related facts are arranged in rows (horizontal) and columns (vertical) so that they can be easily compared and analysed.
A good table is one that is simple, self-explanatory, complete and designed to bring out the desired comparison clearly. The following are the essentials of a good table:
The main components of a table are:
The general rules for constructing a good table are:
Diagrams are visual representations of statistical data that make the data attractive and easy to understand. The main types of diagrams are:
Graphs differ from diagrams in that they are plotted with the help of a coordinate system, with values on the X-axis and Y-axis. The important graphs are:
Histogram: A histogram is a set of adjacent rectangles drawn for a continuous frequency distribution. The class intervals are taken on the X-axis and the frequencies on the Y-axis. For equal class intervals, the area of each rectangle is proportional to the frequency of the class. The histogram represents the frequency distribution graphically.
Frequency polygon: A frequency polygon is obtained by joining the mid-points of the tops of the rectangles of the histogram with straight lines. It can also be drawn without a histogram by plotting the mid-values against the frequencies and joining the points. The polygon is closed by joining it to the base at the mid-value points of the imaginary classes at the two ends.
Ogive (cumulative frequency curve): An ogive is the graph of a cumulative frequency distribution. Two types are drawn: - Less than ogive: Cumulative frequencies 'less than' the upper limits of the classes are plotted against the upper limits. - More than ogive: Cumulative frequencies 'more than' the lower limits are plotted against the lower limits. The point where the two ogives intersect gives the median of the distribution.
Diagrams and graphs are attractive, easy to understand, facilitate comparison and make data memorable. They are particularly useful in presentations and for highlighting trends and patterns. However, they have limitations: they cannot present large or very precise data, they give only an approximate picture, they may be misleading if drawn with false base lines or unsuitable scales, and they are not suitable for further statistical calculations. Hence tables, which present precise figures, must always accompany diagrams and graphs.
| Component of Table | Meaning |
|---|---|
| Table number | Number given for reference |
| Title | Describes what, where, when |
| Caption / Stub head | Column heading / Row heading |
| Head note | Units of measurement |
| Stub | Row headings at the left |
| Body | Numerical data |
| Source note | Where data came from |
| Footnote | Explanation of items |
| Diagram / Graph | Type | Use |
|---|---|---|
| Simple bar diagram | One-dimensional | Single variable over time |
| Multiple bar diagram | One-dimensional | Two or more variables |
| Sub-divided bar | One-dimensional | Components of a total |
| Pie chart | Two-dimensional | Parts of a whole |
| Histogram | Graph | Continuous frequency distribution |
| Frequency polygon | Graph | Mid-values joined by lines |
| Ogive | Graph | Cumulative frequency; gives median |
This chapter dealt with the presentation of data, the link between organised data and statistical analysis. We saw that data may be presented textually, in tables, or in diagrams and graphs. The statistical table, with its components of title, caption, stub, head note, body, source and footnote, presents data precisely and is the backbone of statistical reporting. Diagrams such as simple, multiple, sub-divided and percentage bars, and the pie chart, make data attractive and easy to compare. Graphs such as the histogram, the frequency polygon and the less than and more than ogives provide a graphical picture of the frequency distribution, and the intersection of the two ogives gives the median. With data thus organised and presented, we are now ready to compute the statistical measures of central tendency, dispersion and correlation in the chapters that follow.