Notes on Normal Distribution and Z Scores
Characteristics of a Normal Curve
Bell-shaped distribution of population scores.
The mean is located at the center, with equal spread on both sides.
The curve approaches the horizontal axis but never touches it (asymptotic).
Inflection points occur at +1 and -1 standard deviations from the mean.
Definition of Z Scores
A z score represents how many standard deviations a raw score is above or below the mean.
It enables comparison across different distributions.
Formula: z = \frac{(X - \mu)}{\sigma} where:
$X$ = raw score
$\mu$ = mean of the distribution
$\sigma$ = standard deviation of the distribution
Features of Z Distributions
The mean of z scores is always 0.
The standard deviation of z scores is always 1.
Z scores maintain the shape of the data distribution during transformation.
Using Z Scores with a Normal Curve
To determine:
(a) Percentage of scores below a given raw score (cumulative area).
(b) Percentage of scores above a given raw score.
(c) Percentage of scores between two raw scores (find respective z scores and calculate area).
Understanding Standard Deviations (SD)
Standard deviation represents the dispersion of data from the mean.
Empirical Rule:
68% of scores lie within ±1 SD from the mean.
95% lie within ±2 SD.
99.7% lie within ±3 SD.
Positive SD indicates higher than average scores; negative SD indicates lower than average scores.
Skewness and Kurtosis
Skewness: Measure of asymmetry in a distribution.
Positive Skew: Longer right tail; mean > median.
Negative Skew: Longer left tail; mean < median.
Kurtosis: Describes the peakedness of the distribution.
Leptokurtic: High peak, more outliers around the mean.
Platykurtic: Broad peak, scores are spread out.
Standardized Scores
Standardized scores allow comparison among different distributions.
A derived score represents an individual's position relative to the mean, expressed in standard deviation units.
Common types: T-scores, Stanines, and Sten Scores.
T-score: Mean of 50, SD of 10. Formula: T = (z)(10) + 50
Stanine: Ranges from 1 to 9 with a mean of 5, SD of 2.
Sten: No midpoint, defined by half standard deviations.
Examples of Z-Score Calculations
To evaluate individual performance:
Jinwoo: Math score of 85 (mean 80, SD 5) → z = \frac{(85 - 80)}{5} = 1
Goto: English score of 90 (mean 88, SD 4) → z = \frac{(90 - 88)}{4} = 0.5
Determine percentile ranks using z scores to find areas under the normal curve.
Finding Percentile Ranks and Raw Scores
Given a raw score, calculate the z score and use normal distribution tables to find corresponding percentiles.
For raw score calculations, rearrange the z score formula to retrieve original scores based on percentiles.
Characteristics of a Normal Curve
Bell-shaped Distribution: The normal curve is characterized by its bell shape, which reflects a symmetrical distribution of population scores. This means that the data is evenly distributed around the mean, with equal frequencies occurring above and below it.
Central Position of the Mean: The mean of the distribution is located at the center of the curve, dividing the data into two equal halves. This central position signifies that approximately 50% of the population scores fall below the mean and 50% fall above it.
Asymptotic Nature: The curve approaches the horizontal axis as it extends outward in both directions but never actually touches it. This property indicates that there is always a non-zero probability for extreme values, no matter how far from the mean.
Inflection Points: The normal curve has inflection points at +1 and -1 standard deviations from the mean. These points represent the transition from concavity to convexity and correspond to the percentages of scores that lie within specific ranges of the mean.
Definition of Z Scores
A z score quantifies how many standard deviations a specific raw score is above or below the mean value of its distribution. It serves as a standardized measure that allows for comparison across different data sets, regardless of their original scales.
Formula: The calculation of a z score utilizes the formula: z = \frac{(X - \mu)}{\sigma} where:
$X$ = the individual raw score
$\mu$ = the mean of the distribution
$\sigma$ = the standard deviation of the distribution
Features of Z Distributions
Mean and Standard Deviation: In a z distribution (standard normal distribution), the mean is always 0 and the standard deviation is always 1. This standardized form allows for easier statistical analysis and interpretation of data.
Transformation of Shape: The transformation of raw scores into z scores does not alter the shape of the data distribution; it maintains their relative positions, facilitating comparative analysis across various distributions.
Using Z Scores with a Normal Curve
Z scores can be utilized for various analyses:
(a) Percentage of Scores Below a Given Raw Score: By calculating the cumulative area under the curve for a specific z score, one can determine the percentage of scores that fall below that raw score.
(b) Percentage of Scores Above a Given Raw Score: Similarly, the area above a z score can be computed to ascertain the percentage of scores exceeding that particular raw score.
(c) Percentage of Scores Between Two Raw Scores: To find the percentage of scores that lie between two raw scores, one must calculate the corresponding z scores for each raw score and then determine the area between those two z scores.
Understanding Standard Deviations (SD)
Dispersion Representation: The standard deviation (SD) is a statistical measure that quantifies the dispersion or variability of data points from the mean. A low standard deviation indicates that the scores tend to be close to the mean, while a high standard deviation signifies a wider spread.
Empirical Rule: The empirical rule is a key principle in understanding the distribution of data in a normal curve:
Approximately 68% of scores lie within ±1 standard deviation from the mean.
About 95% of scores fall within ±2 standard deviations.
Nearly 99.7% of scores are contained within ±3 standard deviations.
This rule helps in assessing probabilities and making predictions about data.
Significance of SD Values: A positive standard deviation indicates that a score is higher than the mean, while a negative standard deviation suggests it is lower than the mean.
Skewness and Kurtosis
Skewness: This measures the degree of asymmetry in a distribution.
Positive Skew: This occurs when the distribution has a longer right tail, indicating that the mean is greater than the median.
Negative Skew: This happens when the distribution has a longer left tail, meaning the mean is less than the median.
Kurtosis: This describes the peakedness of the distribution.
Leptokurtic: A distribution with a high peak, suggesting more data points are clustered around the mean with fewer extreme outliers.
Platykurtic: A distribution with a broad peak, indicating that scores are more widely spread across the range.
Standardized Scores
Purpose of Standardized Scores: Standardized scores are essential for comparing scores across different distributions. They express an individual's standing relative to the mean in units of standard deviation, facilitating easier interpretation of performance.
Common Types of Standardized Scores:
T-scores: Standardized scores with a mean of 50 and standard deviation of 10. Formula: T = (z)(10) + 50
Stanine Scores: These range from 1 to 9, with a mean of 5 and standard deviation of 2, providing a simplified view of score distributions.
Sten Scores: Not defined by a midpoint, these scores are categorized using half standard deviations, offering a different method for score interpretation.
Examples of Z-Score Calculations
To exemplify how to evaluate individual performance using z scores:
Jinwoo's Case: Jinwoo scored 85 in Math, with a mean of 80 and a standard deviation of 5. Calculation: z = \frac{(85 - 80)}{5} = 1 This indicates Jinwoo scored 1 standard deviation above the mean.
Goto's Case: Goto scored 90 in English, with a mean of 88 and a standard deviation of 4. Calculation: z = \frac{(90 - 88)}{4} = 0.5 This shows Goto scored half a standard deviation above the mean.
Z scores can be employed to determine percentile ranks by finding areas under the normal curve corresponding to those scores.
Finding Percentile Ranks and Raw Scores
To find percentile ranks based on a raw score, first, calculate the z score, then use standard normal distribution tables to ascertain the corresponding percentiles.
Conversely, to compute raw scores from percentiles, rearrange the z score formula to solve for the original scores based on the obtained percentiles. This is especially useful in educational settings where determining student performance relative to a standard is necessary.