Describing Graphed Distributions

Describing Graphed Distributions

Types of Graphical Representations

  • Stem-and-Leaf Plots: A device for displaying quantitative data where each data value is split into a "stem" (the leading digit) and a "leaf" (the trailing digit).

  • Histograms: A graphical representation showing the frequency distribution of a dataset by using bars.

  • Polygons: A line graph representing the frequency of data points, connected at the midpoints of the histogram's bars.

Aspects of Graphed Distributions

  • Central Tendency: A measure that represents the center or typical value of a dataset. Common measures include mean, median, and mode.

  • Dispersion (Variability): The extent to which data points in a dataset differ from each other and from the mean. Common measures include range, variance, and standard deviation.

  • Skewness: A measure of the asymmetry of the probability distribution.

    • Positive skewness indicates that the tail on the right side of the distribution is longer or fatter than the left.

    • Negative skewness indicates that the tail on the left side is longer or fatter than the right.

  • Kurtosis: Refers to the "peakedness" of a distribution, characterizing the shape of the distribution's tails.

    • Lepotokurtic: Distribution is thin and peaked ($ ext{Pearson kurtosis} > 3$).

    • Mesokurtic: Distribution is moderate in shape ($ ext{Pearson kurtosis} = 3$).

    • Platykurtic: Distribution is flat ($ ext{Pearson kurtosis} < 3$).

  • Modality: Refers to the number of peaks (modes) in the distribution. A distribution can be unimodal, bimodal, or multimodal.

  • Outliers: Data points that differ significantly from other observations, potentially influencing the mean and other statistical measures.

Example Maths Ability Scores

  • Scores: Scaled from 0 to 150.

  • Example Scores: 38, 53, 54, 58, 60, 62, 64, 65, 68, 70, 71, 73, 74, 76, 77, … , 147

Frequency Table of Maths Ability

  • Structure: Displays the frequency ( ext{number of occurrences}) of specific scores alongside their cumulative percentages.

  • Columns in Table:

    • Valid Frequency: The count of data points for given scores.

    • Percent: Percentage representation of the valid frequency in relation to total observations.

    • Cumulative Percent: Running total of the percentages, showing the percentage of scores that fall below a particular value.

Score

Valid Frequency

Percent

Cumulative Percent

38

1

0.6

0.6

53

1

0.6

1.1

54

1

0.6

1.7

147

1

Displaying Distributions Graphically

  • Histograms: Graphical representation that shows the distribution of scores by dividing the data into bins and counting the frequency of scores in each bin.

  • Polygons: Similar to histograms but display the frequency using a connected line across the midpoints of each bin.

The Normal Distribution

  • Characteristics: Most graphed distributions are discussed in relation to the normal distribution, characterized by its bell-shaped curve.

  • Natural Variables: Many naturally occurring variables, such as height, weight, and IQ, tend to follow a normal distribution.

    • Example: Individuals with extremely high or low scores are rare, whereas most people cluster around the average.

Description of Graphed Distributions

When describing a graphed distribution, refer to the following:

  • Kurtosis: Assess the peakedness (or flatness) of the graph.

  • Dispersion: Consider the spread or variability of the data.

  • Presence of Outliers: Evaluate if any extreme values deviate significantly from the rest.

  • Skewness: Consider if the distribution leans to one side (positive or negative skew).

  • Modality: Identify the number of peaks in the distribution.

Kurtosis Explained

  • Kurtosis Definition: Refers to how tall and sharp the peaks of a distribution are compared to a normal distribution.

    • Lepotokurtic Distribution: Tall and thin; $ ext{Pearson kurtosis} > 3$.

    • Mesokurtic Distribution: Moderately peaked; $ ext{Pearson kurtosis} = 3$.

    • Platykurtic Distribution: Flat; $ ext{Pearson kurtosis} < 3$.

Skewness Explained

  • Definition of Skewness: The degree of asymmetry of a distribution.

    • A positively skewed distribution: The tail on the right is longer; $ ext{mean} > ext{median} > ext{mode}$.

    • A negatively skewed distribution: The tail on the left is longer; $ ext{mean} < ext{median} < ext{mode}$.

Conclusion on Distribution Characteristics

To effectively describe the distributions observed in data:

  • Assess the central tendency, variability, presence of outliers, kurtosis, skewness, and modality.