PPT for CHAPTER 3 for Civil&arc
Measures of Central Tendency
Definition
Central tendency summarizes a data set into a single value around which other values cluster.
Aims to condense data for comparison through averages.
Objectives
Comprehend data easily.
Provide a single representative value for the data set.
Summarize/reduce data volume.
Facilitate group comparisons.
Enable further statistical analysis.
Summation Notation
Symbol 'Σ' denotes the sum of data values.
Given 'n' observations X1, X2, ..., Xn, the notation represents the sum of all measurements.
Properties of Summation Notation
Constant multiple factor (C) can be factored out: ΣC = nC
Product of a constant and a variable: Σ(bX) = bΣ(X)
Linear combination: Σ(aX + b) = aΣ(X) + nb
Distributive property: Σ(X + Y) = ΣX + ΣY
Characteristics of a Good Measure
Not affected by extreme values.
Unique measurement exists.
Comprehensive as it considers all observations.
Easy to compute and interpret.
Defined rigidity.
Sampling stability.
Usable for further analysis.
Types of Measures of Central Tendency
Mean
Four types: Arithmetic, Weighted, Geometric, and Harmonic.
Arithmetic Mean:
Sum of values divided by count of values.
Different formulas for row data and discrete/continuous frequency distributions.
Median
Middle value when arranged in order; important for both ungrouped and grouped data.
Formula for grouped data requires interpolating class boundaries.
Mode
Value that occurs most frequently.
Can be uni-modal, bi-modal, or multi-modal.
Formula for grouped data considers class frequency.
Midrange
Average of highest and lowest values: MR = (Lowest + Highest) / 2.
Quantiles
Divide data into equal parts: Quartiles, Deciles, Percentiles.
Formulas vary for ungrouped and grouped data.
Example Calculations
Mean for Monthly Income Example
Use formulas for both ungrouped and grouped distributions.
Median Calculation
Easier interpretation for skewed distributions compared to mean.
Mode Calculation for Grouped Data
Identifying modal classes based on frequencies.
Comparison of Measures
Median is less affected by skewed data and extreme values compared to the mean.
Mode is the simplest measure and irrelevant for data distributions lacking a mode.
Midrange is rough and highly susceptible to outliers.
The effectiveness of each measure depends on the data's nature and distribution.
Conclusion
Understanding the right measure of central tendency enhances data analysis and interpretation.
Each measure has its advantages and limitations; selection should depend on the data set characteristics.