Comprehensive Notes on Statistical Variables, Sampling, and Numerical Measures

Overview of Population, Sample, Parameter, and Statistic

Definition of Population:
- A population in statistics represents the entire set of individuals, items, or events being studied.
- A population does not need to be exceptionally large; it can be restricted to a specific small group depending on the research scope (e.g., all students present in a specific classroom).
Population Parameter vs. Sample Statistic:
- Population Parameter: A numerical descriptive measure of an entire population. In practice, true population parameters are rarely known because evaluating an entire population is often impossible or impractical.
- Sample Statistic: A numerical descriptive measure calculated from a sample, used directly to estimate an unknown population parameter.
- Role of Statistics in Estimation: Rather than leaving an unknown parameter as an unstated variable $x\%$ , a sample statistic assigns a concrete, numerical value to serve as the point estimate for the parameter.
Demonstration of Sampling and Estimation:
- Population Parameter Setup: In a classroom population, the exact percentage of students whose favorite color is pink is known to be $18\%$ (clarified from an initial estimate of $26\%$ ).
- Random Sample Collection: A random sample of $3$ students is selected from the classroom population, yielding the following favorite colors:
- Student 1: Pink
- Student 2: Blue
- Student 3: Black
- Sample Statistic Calculation: Out of the $3$ sampled students, $1$ preferred pink. The resulting sample statistic for the proportion of students who prefer pink is: $\frac{1}{3} \approx 33.33\%$
- Application of Estimates: If the population parameter of $18\%$ were unknown, the statistical process would yield $33\%$ as the best working estimate of the population parameter based strictly on the collected sample data.
The Four-Step Statistical Process:
- Step 1: Set the Population: Formally define the target population for the study (e.g., all students in a classroom).
- Step 2: Collect a Sample: Select a subset of individuals from the defined population out of the multiple possible samples that could be drawn.
- Step 3: Estimate the Parameter: Calculate the sample statistic from the gathered sample data to estimate the unknown population parameter.
- Step 4: Conclude the Study: Conclude the investigation using the calculated statistic as the representative estimate.

Classification of Statistical Variables: Qualitative vs. Quantitative

Statistical Variables vs. Algebraic Variables:
- In algebra, variables such as $x$ , $y$ , or $z$ act as placeholders that can be substituted with any numerical value within their mathematical domain.
- In statistics, variables represent distinct characteristics or attributes measured across individuals within a population.
Definition of an Individual:
- An individual is any single element, entity, or member belonging to a defined population (e.g., a single student within a classroom population).
Primary Categories of Statistical Variables:
- Qualitative Variables (Categorical Variables):
- Allow for the classification of individuals based on some characteristic, quality, or attribute.
- They place individuals into specific non-numerical categories.
- Quantitative Variables:
- Provide numerical measures of individuals.
- Operations such as addition or averaging are mathematically meaningful when applied to quantitative variables.
Examples and Classification Practice:
- Temperature:
- Classed as Quantitative.
- Temperature represents a physical measurement that yields explicit numerical values (e.g., $70\,^{\circ}\text{F}$ , $90\,^{\circ}\text{F}$ ).
- Conceptual Nuance: If temperature is strictly categorized into terms like "hot" or "cold", it functions qualitatively; however, standard numerical measurements of temperature are quantitative.
- Gender:
- Classed as Qualitative.
- It assigns individuals to non-numerical categories (e.g., male or female).
- ZIP Code:
- Classed as Qualitative.
- Although ZIP codes consist of digits, they represent geographical locations and administrative labels rather than numerical measurements. Performing arithmetic operations on ZIP codes yields no meaningful mathematical result (analogous to numbers on an athlete's sports jersey).
- Number of Hours a Student Studied for an Exam:
- Classed as Quantitative.
- It measures an amount of time that can be quantified numerically.

Sub-Types of Quantitative Variables: Discrete vs. Continuous

Classification Hierarchy:
- Statistical variables branch into qualitative and quantitative types. Quantitative variables further divide into discrete and continuous variables.
Definitions and Distinctions:
- Discrete Variable:
- A quantitative variable that has a countable number of possible values.
- Associated with values obtained by counting discrete units (typically whole numbers, using one's fingers or counters).
- Continuous Variable:
- A quantitative variable that has an infinite number of possible values over a continuous spectrum.
- Associated with values obtained by measuring a continuous domain (including fractions and decimals).
Examples of Discrete vs. Continuous Variables:
- Number of heads obtained after flipping a coin $5$ times:
- Classed as Discrete.
- The outcome is countable, with possible distinct values belonging to the set $\left\{0, 1, 2, 3, 4, 5\right\}$ .
- Number of cars arriving at a McDonald's drive-through between $12\,\text{PM}$ and $1\,\text{PM}$ :
- Classed as Discrete.
- The number of cars is determined by counting whole units (e.g., $0, 1, 2, 3$ cars). Fractional values, such as $3.75$ cars, cannot occur.
- Distance a Tesla Model Y can travel in one hour:
- Classed as Continuous.
- Distance cannot be counted; it must be measured along a continuous scale.
- The result can take an infinite number of real numerical values depending on driving conditions (e.g., $52\,\text{miles}$ , $73.5\,\text{miles}$ ).

Identifying Population and Sample in Practical Studies

Requirement for Specificity:
- Populations and samples must be specified with precise detail, including age ranges, temporal boundaries, geographical locations, and target measurements, to prevent ambiguity.
Case Study 1: Teenager Mental Health Prescriptions:
- Study Context: The Gallup organization surveyed $1,028$ teenagers aged $13$ to $17$ years living in the United States to determine whether they had been prescribed medications for any mental disorders.
- Identified Population: All teenagers aged $13$ to $17$ living in the United States (or all potential responses from teenagers aged $13$ to $17$ in the United States).
- Identified Sample: The specific $1,028$ teenagers aged $13$ to $17$ living in the United States who were surveyed (or their $1,028$ collected responses).
- Evaluation of Sample Representativeness:
- Potential Flaws / Limitations:
  1. Sample Size Constraints: Evaluating only $1,028$ teenagers represents a relatively small sample relative to the millions of teenagers in the United States.
  2. Age Exclusion: The study design excludes $18$ -year-olds (and $19$ -year-olds), who encounter distinct legal, financial, and adult responsibilities that could alter prescription rates.
- Practical Trade-offs: Achieving perfect representative samples is restricted by time and monetary constraints. Increasing sample size directly increases logistical costs.
Case Study 2: Beverage Manufacturing Quality Control:
- Study Context: A quality control manager randomly selects $50$ bottles of Coca-Cola filled on October 15 to evaluate the calibration of the filling machine.
- Identified Population: All bottles of Coca-Cola filled on October 15 (specific to the calibration assessment period of that target date).
- Identified Sample: The $50$ bottles of Coca-Cola filled on October 15 that were randomly selected.
- Methodological Importance of Randomization: Random selection within the sample design prevents systematic bias and yields an accurate representation of the population.
Case Study 3: Agricultural Crop Yield Measurement:
- Study Context: A farmer interested in evaluating the weight of his soybean crop randomly samples $100$ plants and weighs the soybeans produced on each plant.
- Abstract Focus of Population/Sample:
- In many studies, the focus of the population is not the physical entities (the plants themselves), but the numerical measurements derived from them (the weights).
- Identified Population: The weights of all soybean crops produced on the farmer's land.
- Identified Sample: The weights of the $100$ randomly selected soybean plants from the farmer's land.
- Real-World Application: Agricultural practices (such as farming in Colquitt County, Georgia) rely on random sampling because weighing every individual crop across large farmland acreage is physically impossible.