AP Statistics Unit 1: Exploring One Variable Data

Fundament Statistics and Data Definitions

 - Individuals and Variables   - Individuals are the objects described by a set of data. These can be anything: people, animals (like lemurs), objects (chairs), biological entities (trees), geographic locations (lakes, states, countries), or even units of time (days).   - A variable is any characteristic of an individual that can change or vary from one individual to another. The study of statistics is necessary specifically because individuals vary.  - Statistics vs. Parameters   - Statistic: Any summary information learned or calculated from a sample of data. A helpful mnemonic is that Statistic and Sample both start with "S."   - Parameter: Any summary information that describes an entire population. A helpful mnemonic is that Parameter and Population both start with "P."  - Types of Variables   - Categorical Variables: These take on values that are category names or group labels. Examples include eye color, hair color, or species type. Values are typically words.    - Exception: Zip Code. Although it is a number, it is categorical because it is not measured or counted; it serves as a label to put mail into a specific city/region category.   - Quantitative Variables: These take on numerical values that represent a measurement or a count. Examples include the weight of a frog or the number of candies in a bag.

Categorical Data Analysis

 - Organizing Categorical Data   - Frequency Table: A table that lists each category and the count (frequency) of individuals that fall into that category. For example, a sample of $89$ lemurs might be categorized by type: Sifaka, Aye-aye, Ring-tail, or Mouse lemur.   - Relative Frequency Table: Instead of counts, this lists the proportion, percentage, or rate of data in each category. Proportion is calculated as: $\frac{\text{count in category}}{\text{total n}}$.  - Graphical Representations   - Bar Graphs: Used to display frequencies or relative frequencies. The x-axis lists categories and the y-axis shows the count or proportion.   - Pie Charts (Circle Graphs): Shows the distribution of categorical data as slices of a whole circle. Each slice represents the proportion of that category relative to the entire set.  - Comparing Categorical Distributions   - When comparing two samples of different sizes, relative frequencies (proportions) should be used rather than raw counts to ensure a fair comparison.   - Describing a categorical distribution involves identifying which categories have the most or least data.

Quantitative Data Analysis and Graphing

 - Subtypes of Quantitative Variables   - Discrete Quantitative Variables: Values that are countable and finite. Typically represent whole numbers, such as the number of goals scored in a soccer game ($0, 1, 2, 3…$).   - Continuous Quantitative Variables: Values that are not countable and can theoretically be infinite within a range. They often involve decimals and depend on the precision of the measuring tool. Example: the weight of a frog ($5.4321\,lbs$).  - Organizing Quantitative Data   - Frequency Table for Numbers: Requires the creation of "bins" or "intervals" of equal size.   - Left-handed Bins: A common convention where a value falls into a bin if it is equal to the left-hand number but less than the right-hand number. For example, a bin for $20$ to $30\,feet$ includes a tree that is exactly $20\,feet$ up to $29.999…\,feet$. A $30\,foot$ tree goes into the next bin.  - Common Quantitative Graphs   - Dot Plot: Places a dot for each individual data value along a number line.   - Stem-and-Leaf Plot: Organizes data by "stems" (leading digits) and "leaves" (trailing digits). This allows the viewer to see every individual data value while still observing the overall shape.   - Histogram: The preferred graph for quantitative data. Bars represent bins of data. The height represents the frequency or relative frequency. Note: Do not confuse a histogram with a bar graph; bar graphs are for categories, histograms are for quantitative intervals.   - Cumulative Graph: Every point $(x, y)$ on the graph indicates that a proportion $y$ of the data is at or below the value $x$.    - Interpretation: A steeper slope indicates a high frequency of data in that range. A horizontal line indicates a gap in the data (no data in that bin).

Describing the Distribution of a Quantitative Variable

 - The Four Essential Features (SCSO)   - Shape: Identify the visual patterns in the data.    - Symmetric: Both sides of the center look roughly the same.    - Unimodal: One clear peak.    - Bimodal: Two clear peaks (may indicate two distinct clusters in the data).    - Uniform: Data is spread evenly across the entire range.    - Skewed Right: The tail of the data extends to higher values (the right side).    - Skewed Left: The tail of the data extends to lower values (the left side).    - Gaps/Clusters: Mention if there are significant empty spaces or distinct groups.   - Center: A single value that summarizes the "typical" entry. This is usually the Mean or Median.   - Spread: How much the data varies. Measured by Range, IQR, or Standard Deviation.   - Outliers/Unusual Features: Mention specific values that depart significantly from the rest of the distribution.

Measures of Center

 - The Mean ($\bar{x}$)   - Calculated by summing all values and dividing by the total count ($n$).   - Formula: $\bar{x} = \frac{\sum x_i}{n}$.   - Sensitivity: The mean is non-resistant or sensitive to outliers. A large outlier will pull the mean toward it to balance the data set.  - The Median ($M$)   - The middle value when data is ordered from least to greatest.   - Finding Location: Use the formula $\frac{n + 1}{2}$ to find the position of the median.    - If $n = 19$, the median is the $10th$ value ($\frac{19+1}{2} = 10$).    - If $n = 20$, the median is between the $10th$ and $11th$ values ($\frac{21}{2} = 10.5$).   - Resistance: The median is resistant to outliers. Extreme values at the ends do not change the middle value much.  - Relationship Between Shape, Mean, and Median   - Symmetric: Mean $\approx$ Median.   - Skewed Left: Mean < Median. Extreme low values pull the mean down.   - Skewed Right: Mean > Median. Extreme high values pull the mean up.

Measures of Position

 - Percentiles   - The $p$-th percentile is the value such that $p$ percent of the data is at or below that value. For example, the $95th$ percentile on the SAT means $95\%$ of students scored at or below your score.  - Quartiles   - First Quartile ($Q1$): The $25th$ percentile; the median of the bottom half of the data.   - Median ($Q2$): The $50th$ percentile.   - Third Quartile ($Q3$): The $75th$ percentile; the median of the top half of the data.

Measures of Spread

 - Range: The absolute difference between the Maximum and Minimum values ($\text{Max} - \text{Min}$). Highly sensitive to outliers.  - Interquartile Range (IQR): The range of the middle $50\%$ of the data. Formula: $IQR = Q3 - Q1$. This is a resistant measure of spread.  - Standard Deviation ($s_x$ or $\sigma$): Measures the average or typical distance data points are from the mean.   - A high standard deviation means data is spread far from the mean.   - A low standard deviation means data is clustered closely around the mean.   - Formula: $s_x = \sqrt{\frac{\sum(x_i - \bar{x})^2}{n-1}}$.

Identifying Outliers: The Official Methods

 - The Fence Method (Quartile-based)   - Upper Fence: $Q3 + 1.5 \times IQR$. Any value greater than this is an outlier.   - Lower Fence: $Q1 - 1.5 \times IQR$. Any value less than this is an outlier.  - The Standard Deviation Method   - Outliers are identified as any value more than two standard deviations away from the mean ($\bar{x} \pm 2s_x$).

Data Transformations

 - Adding or Subtracting a Constant   - Affects: Measures of Center (Mean/Median) and Position (Quartiles/Percentiles). Everything shifts up or down by that constant.   - Does NOT affect: Measures of Spread (Range, IQR, Standard Deviation).  - Multiplying or Dividing by a Constant   - Affects: Everything. Measures of Center, Position, and Spread are all multiplied/divided by that constant.  - Changing Specific Data Points   - Adding a high outlier: The mean increases significantly; the median moves very little (if at all).   - Adding a value near the middle: Both the mean and median remain relatively stable.

Box Plots and the Five-Number Summary

 - Five-Number Summary: Includes Minimum, $Q1$, Median, $Q3$, and Maximum.  - Box Plot Construction   - A box is drawn from $Q1$ to $Q3$ with a line for the Median.   - Modified Box Plot: Outliers are identified via the fence method and plotted as asterisks ($*$). Whiskers extend only to the next highest/lowest non-outlier data point.  - Interpretation: Each of the four sections of the box plot (whisker, box half, box half, whisker) represents exactly $25\%$ of the data. A wider section does not mean more data; it means that $25\%$ of the data is more spread out.

The Normal Distribution

 - Density Curves: Models used to describe the distribution of a population. The total area under any density curve is exactly $1$.  - Properties of the Normal Distribution   - Shaped: Symmetric, bell-shaped (mound-shaped), and unimodal.   - Parameters: Defined by the population mean ($\mu$) and population standard deviation ($\sigma$).   - Note: The normal model applies to continuous quantitative variables.  - The Empirical Rule (68-95-99.7 Rule)   - Approximately $68\%$ of data falls within $1\sigma$ of $\mu$.   - Approximately $95\%$ of data falls within $2\sigma$ of $\mu$.   - Approximately $99.7\%$ of data falls within $3\sigma$ of $\mu$.

Standardized Scores (Z-Scores)

 - Definition: A Z-score measures exactly how many standard deviations a value falls above or below the mean.  - Formula: $z = \frac{x - \mu}{\sigma}$  - Purpose: Standardizing allows for the comparison of values from different distributions (e.g., comparing the height of a tree in feet to the weight of a bear in pounds).  - Standard Normal Model: A normal distribution with $\mu = 0$ and $\sigma = 1$.

# Normal Distribution Calculations  - Finding Proportions/Areas   - Proportions below: Use technology or a Z-table. On TI-84, use normalcdf(lower: -99, upper: z, mean: 0, sd: 1).   - Proportions above: On TI-84, use normalcdf(lower: z, upper: 99, mean: 0, sd: 1). If using a Z-table, calculate $1 - \text{area below}$.   - Proportions between: On TI-84, use normalcdf(lower: z1, upper: z2, mean: 0, sd: 1).  - Working Backwards (Inverse Normal)   - Used when given a percentile or area to find the corresponding Z-score or raw value ($x$).   - Technology: On TI-84, use invNorm(area: proportion_to_left, mean: 0, sd: 1).   - Example: To find the height at the $80th$ percentile for a tree population ($\mu = 80, \sigma = 18$):    1. Find $z$ for the $0.80$ area: $z \approx 0.842$.    2. Solve for $x$: $0.842 = \frac{x - 80}{18}$.    3. $x = (0.842 \times 18) + 80 = 95.156\,feet$.