These answers are written in an exam-oriented style for IOE engineering students. Read the question, think of your own answer, then click Check Answer.
Measures of central tendency are numerical values that represent the centre or typical value of a data set. They give a single representative value around which the observations tend to cluster.
Chief measures of central tendency:
Techniques to measure consistency:
Consistency means the degree of uniformity or stability in the data. A data set with less variation is said to be more consistent. Consistency is studied using measures of dispersion such as range, variance, standard deviation and coefficient of variation.
\[ CV=\frac{S.D.}{Mean}\times100 \]
For comparing two or more data sets, coefficient of variation is preferred. The data set with smaller coefficient of variation is more consistent.
Measures of central tendency and measures of dispersion are two important descriptive statistical tools. Measures of central tendency describe the central position of the data, whereas measures of dispersion describe the amount of variation present in the data.
| Measures of Central Tendency | Measures of Dispersion |
|---|---|
| They provide a single representative value for the whole data set. | They describe the spread, scatter or variability of the observations. |
| They indicate the location around which the observations are concentrated. | They indicate the extent to which the observations differ from one another. |
| Examples are mean, median and mode. | Examples are range, variance, standard deviation and coefficient of variation. |
| They are useful for summarising the general level of the data. | They are useful for judging uniformity, reliability and consistency. |
For example, two construction materials may have the same mean compressive strength. However, the material having smaller standard deviation or smaller coefficient of variation is considered more consistent and reliable.
Statistics is highly significant in engineering because engineering decisions are often based on data collected from experiments, field measurements, production processes and real systems. It helps engineers analyse data, understand variation and make reliable decisions under uncertainty.
Significance of statistics in engineering:
For example, a civil engineer may use statistics to analyse concrete strength, rainfall data, traffic volume, soil properties or quality variation in construction materials.
Median is a positional average. It is the value that divides an ordered data set into two equal parts. Half of the observations lie below it and half lie above it.
Merits of median:
Demerits of median:
Thus, median is very useful when extreme observations are present, but it is less useful for advanced mathematical calculations.
A pie diagram is a circular diagram used to represent the division of a total quantity into different parts. The whole circle represents the total, and each sector represents a component of the total.
Method of construction:
\[ \text{Sector angle}=\frac{\text{Component value}}{\text{Total value}}\times360^\circ \]
Pie diagrams are useful when the objective is to compare the relative contribution of each part to the whole.
Graphical presentation is the method of representing numerical data using diagrams and graphs. It helps readers understand the nature of data quickly and clearly.
Importance:
Limitations:
Graphical methods:
A measure of central tendency is a single value that represents the centre or average position of a data set. It gives a summary value around which the observations are distributed.
The common measures of central tendency are arithmetic mean, median and mode.
Requisites of a good measure of central tendency:
No single average is best in all situations. The arithmetic mean, median or mode should be selected according to the nature of the data and the purpose of analysis.
Measures of dispersion describe the spread or variability of observations in a data set. They may be classified into absolute measures and relative measures.
Absolute measures of dispersion:
Absolute measures express variation in the same unit as the original data. They are useful for studying variation within a single data set.
Relative measures of dispersion:
Relative measures express variation as a ratio or percentage. They are independent of units and are useful for comparing two or more data sets.
\[ CV=\frac{S.D.}{Mean}\times100 \]
If two data sets have different units or different means, coefficient of variation is more appropriate for comparison. The data set with smaller CV is more consistent.