Introduction to Statistics

Definition: Statistics can be categorized into two main types:
- Descriptive Statistics:
  - Focuses on summarizing and describing the features of a dataset.
  - Example: Calculating the average height of students in a class.
- Inferential Statistics:
  - Takes descriptive data and uses it to infer or predict characteristics about a larger population based on a sample.
  - Example: If the average height of a class is 6 foot 4, one might infer that "students at Quinnipiac are generally tall" based on this sample.

Categories: Statistics relies on four main types of data:
- Nominal:
  - Definitions and Characteristics: Data that consists of names or categories without any intrinsic ordering.
  - Example: Shirt color (e.g., red, blue, green, black).
- Ordinal:
  - Definitions and Characteristics: Data with a meaningful order, but the intervals between values are not necessarily equal.
  - Example: Rating of a restaurant (i.e., Poor, Fair, Good, Very Good, Excellent).
  - Explanation: There is a clear order (e.g., poor is less than fair).
- Interval:
  - Definitions and Characteristics: Numeric data with meaningful intervals between values but no true zero point.
  - Example: Temperature measured in degrees Fahrenheit.
  - Explanation: 0 degrees Fahrenheit doesn’t indicate an absence of temperature.
  - Note: Continuous data because it can be represented as decimals (e.g., 98.6 F).
- Ratio:
  - Definitions and Characteristics: Numeric data with a true zero point, allowing for comparisons of absolute magnitudes.
  - Example: Number of questions answered correctly on a test (out of 50).
  - Explanation: 0 questions correct means nothing correct at all.
  - Note: Discrete data because you can only have integers.

Examples:
- Brand of smartphone (e.g., Apple, Samsung, Google) - Nominal.
- College major (e.g., Biology, Chemistry, Computer Science) - Nominal.
- Time spent studying for an exam (in hours) - Ratio:
  - Explanation: 0 hours means no study, and time can be decimal (e.g., 1.5 hours).
- Number of absences from class during the semester - Ratio:
  - Explanation: 0 absences indicate perfect attendance; data is discrete (e.g., cannot have 1.5 absences).

Types of Survey Methods:
1. Simple Random Sampling:
  - Definition: Every individual in the sample has an equal chance of being chosen.
2. Systematic Sampling:
  - Definition: Selects every kth person in the population (e.g., every 8th person).
3. Convenience Sampling:
  - Definition: Collecting data from individuals who are easiest to reach (e.g., surveying people passing by).
4. Stratified Random Sampling:
  - Definition: Picks a few individuals from all groups within a population.
  - Example: From eight groups of 100, select 20 from each group.
5. Cluster Sampling:
  - Definition: Collecting data on everyone within selected groups.
  - Example: Sampling every person in 1 or 2 out of 5 groups.

Types of Skewness:
- Left-Skewed:
  - Definition: The tail of the distribution extends to the left.
- Right-Skewed:
  - Definition: The tail of the distribution extends to the right.
- Symmetric:
  - Definition: Distribution is balanced; left and right sides are mirror images.

Mean vs. Median in different skewness scenarios:
- Left-Skewed: Mean < Median; pulls down smaller values.
- Right-Skewed: Mean > Median; pulls up larger values.

Types:
1. Range:
  - Definition: Difference between the highest and lowest values.
  - Least Accurate:
    - Reason: Only uses two numbers, which could be outliers, providing little insight into the dataset's overall distribution.
2. Variance:
  - Definition: A measure of how far each number in the dataset is from the mean, squared.
  - Usage: Primarily theoretical due to squared units.
3. Standard Deviation:
  - Definition: The square root of variance, giving a measure of dispersion in the same units as the data.
  - Usage: Preferable for practical applications, as it retains original units.

Empirical Rule:
- Definition: In a normal distribution, approximately 68% of data points fall within ±1 standard deviation from the mean.
Chebyshev’s Inequality:
- Application: Can be used for any distribution, providing lower bounds for the proportion of data within certain standard deviations.
- Example: For any distribution:
  - At least 75% of data is within ±2 standard deviations.
  - At least 88.89% is within ±3 standard deviations.

Problem Example (Heights of American Men):
- Given: Mean = 70 inches, Standard Deviation = 2.5 inches.
- Questions:
  - What percentage of men are between 65 inches and 75 inches?
  - Implication: Utilize the Empirical Rule for bell-shaped distributions.
  - More challenging: What percentage are between 65 inches and 72.5 inches?

Statistical analysis in Excel:
- Sample regression formula:
  - $y = -0.1713X + 254.95$
- R-squared value:
  - 0.8969
- R value:
  - -0.947 (consistent with negative relationship noted in scatterplot).
- Mean value for exercise minutes column:
  - 179.87 minutes