Measure of Central Tendency
Topic 3: Measure of Central Tendency
Represents the center of a data set.
A single summary number indicates where many of the scores lie.
Types of Measures of Central Tendency
A. Mean
The arithmetic average of a data set.
Formula: Mean = Σx/n (sum of all scores divided by number of scores)
Example 1:
Scores: 1.42, 1.97, 1.42, 1.50, 1.67
Mean = 7.98 / 5 = 1.60
Example 2:
Scores: 2.00, 3.40, 7.00, 11.00, 23.00, 3.41
Calculation needed for Mean.
Can also be calculated using statistical software like Excel.
B. Median (mdn)
The midpoint of a data set.
Divides values into two equal parts, locating the middle value.
Calculation: Location of Median (L) = (n + 1) / 2
Example 1:
Scores: 11, 12, 13, 14, 15
L = (5 + 1) / 2 = 3 → Median = 13
Example 2:
Scores: 2, 4, 4, 6, 8, 8
L = (6 + 1) / 2 = 3.5 → Median = (4 + 6) / 2 = 5
C. Mode
The most frequently occurring value in a data set.
Example 1:
Scores: 1, 7, 5, 9, 8, 7 → Mode = 7
Example 2:
Scores: 1, 7, 5, 9, 8 → No Mode
Example 3:
Scores: 1, 7, 7, 5, 9, 9, 8 → Bimodal (Modes = 7 and 9)
Mode is primarily useful for categorical data or large datasets of interval/ratio data.
When to Use Each Measure
Nominal Data: Use Mode (mean and median not applicable).
Example: Law = 64; Kine = 59; Eng. = 37
Ordinal Data: Use Median.
Example: Positional rankings (1st, 2nd, 3rd, etc.).
Interval or Ratio Data: Use Mean and/or Median.
Use Median if data is highly skewed or has outliers (as it's less affected).
Example of skew:
Player salaries → Mean = $495,000; Median = $125,000 due to outlier.
Excel: Calculating Central Tendency Measures
Excel Formulas:
Mean: = average(data_set)
Median: = median(data_set)
Mode: = mode(data_set)
Descriptive Statistics can be run using Toolpak in Excel to provide detailed statistics including mean, median, mode, range, standard deviation, and more.
Graphical Representation of Data
Summary data can be graphically depicted using:
Bar Graphs: Represent means or medians with categories on the x-axis.
Histograms: Display the frequency distribution of interval/ratio data.
Types of Data Distributions
Symmetrical Distribution: Frequencies decrease evenly on both sides of the center (e.g., mean = median = mode).
Skewed Distribution: Frequencies bunched at one end of the scale.
Positively Skewed: Bunched at lower scores; mean > median > mode.
Negatively Skewed: Bunched at higher scores; mode > median > mean.
Variability and Kurtosis
Spread: Refers to how much data varies.
Kurtosis: Indicates the peakness or flatness of a distribution.
Leptokurtic: More peaked, fewer scores in tails.
Platykurtic: Flatter distribution, more evenly spread out.
Importance of Graphing
Reliance solely on summary statistics can be misleading. Always visualize data through graphs to understand its distribution.