Random Variables and Sample Survey Fundamentals
Random Variables, Expectation, Variance, and Covariance
Definition of Random Variable
A random variable is a function that assigns a real number to each outcome in the sample space of a random experiment. It bridges the gap between abstract experimental outcomes and numerical analysis.
Definition: Let S be a sample space. A random variable X is a real-valued function X: S → ℝ such that the probability of each subset of outcomes is well-defined.
Random variables are broadly classified into two types:
- Discrete Random Variable: Takes a countable number of distinct values (e.g., number of defective items in a batch).
- Continuous Random Variable: Takes an uncountable infinite number of values within a given interval (e.g., exact height or weight of students).
Expectation (Mathematical Expectation)
The expectation, or expected value, represents the theoretical long-run average or mean value of a random variable over numerous trials.
Formula (Discrete): E(X) = ∑ xi pi
Formula (Continuous): E(X) = ∫ x f(x) dx
Practical Example: If a game pays 10 dollars with a probability of 0.2 and 0 dollars with a probability of 0.8, the expected value is E(X) = (10 × 0.2) + (0 × 0.8) = 2 dollars.
Variance
Variance measures the dispersion or spread of a random variable around its expected value. It quantifies how much the values deviate from the mean.
Formula: Var(X) = E([X - E(X)]2) = E(X2) - [E(X)]2
Exam-Oriented Note: Variance is always non-negative. Standard deviation is the positive square root of variance, denoted as σ = √Var(X).
Covariance
Covariance measures the joint variability of two random variables. It indicates the direction of the linear relationship between variables.
Formula: Cov(X, Y) = E(XY) - E(X)E(Y)
- Positive Covariance: When X increases, Y tends to increase as well.
- Negative Covariance: When X increases, Y tends to decrease.
- Zero Covariance: Indicates no linear relationship between X and Y.
Addition and Multiplication Theorems of Expectation
These theorems simplify the calculation of expectations for sums and products of random variables.
- Addition Theorem of Expectation: For any two random variables X and Y, the expected value of their sum is the sum of their individual expected values, regardless of whether they are independent.
E(X + Y) = E(X) + E(Y)
- Multiplication Theorem of Expectation: If X and Y are independent random variables, the expected value of their product is the product of their individual expected values.
E(XY) = E(X) × E(Y) (for independent variables)
| Theorem | Condition | Mathematical Statement |
|---|---|---|
| Addition Theorem | Always holds (Any X, Y) | E(X + Y) = E(X) + E(Y) |
| Multiplication Theorem | X and Y must be independent | E(XY) = E(X)E(Y) |
Population, Sampling, and Errors
Concept of Population and Sample
In statistics, understanding the distinction between the entire group of interest and a subset used for study is fundamental.
- Population: The complete set of all elements, individuals, or observations sharing a common characteristic under study.
- Sample: A representative subset drawn from the population, used to infer population characteristics.
Complete Enumeration versus Sampling
Data collection can be performed by studying every single member of a population or by studying a representative sample.
| Feature | Complete Enumeration (Census) | Sampling |
|---|---|---|
| Definition | Collection of data from every unit in the population. | Collection of data from a subset of the population. |
| Cost and Time | Extremely high cost and time-consuming. | Cost-effective and rapid data collection. |
| Accuracy | High accuracy theoretically, but prone to large operational errors due to scale. | High accuracy when properly designed, minimizing operational fatigue. |
| Applicability | Best for small populations or mandatory national counts. | Essential for large, infinite, or destructive testing populations. |
Sampling and Non-Sampling Errors
Data collected through surveys is susceptible to various errors that impact the reliability of results.
- Sampling Errors: Errors that arise because data is collected from a sample rather than the entire population. It is due to natural variation between sample estimates and population parameters. Can be reduced by increasing sample size.
- Non-Sampling Errors: Errors that occur in both census and sample surveys due to flaws in data collection, processing, or response. Examples include non-response bias, measurement errors, and interviewer bias.
Survey Methodology and Pilot Surveys
Pilot Survey
A pilot survey is a small-scale preliminary study conducted before the main survey to test questionnaire design, field logistics, and estimation procedures.
- Purpose: Identifies unforeseen practical difficulties, checks the clarity of questions, and estimates variance for determining optimal sample size.
- Real-World Application: Market researchers run a pilot test on 50 consumers before launching a national survey of 10,000 consumers.
Basic Principle of Sample Survey
Scientific sample surveys rely on core statistical principles to ensure valid inferences about the population.
- Randomization: Every unit in the population must have a known, non-zero probability of being selected. This eliminates selection bias.
- Optimization: Achieving maximum precision at a minimum cost, or achieving desired precision with the minimum sample size.
- Validity: Ensuring that the sample accurately reflects the structure and characteristics of the target population.
Common Mistake: Confusing convenience sampling with random sampling. Convenience sampling violates the basic principle of sample surveys because not every unit has a known chance of selection, leading to biased estimates.