Chapter 1: Introduction to Statistics – Key Concepts and Methods

Introduction to Statistics

What is Statistics?

Statistics is the science of collecting, organizing, analyzing, and interpreting data to draw conclusions and make informed decisions.

• Collecting data: Gathering information through measurements, surveys, or experiments.

• Organizing & summarizing data: Using tables, charts, and graphs to present data clearly.

• Analyzing data: Applying statistical methods to extract meaningful patterns.

• Interpreting results: Drawing logical conclusions based on data analysis.

The Statistical Process

Three Main Steps

1. Prepare: Define the context, goals, and data collection methods.

2. Analyze: Use graphs, charts, and statistical techniques to examine data.

3. Conclude: Interpret results, assess significance, and draw conclusions.

Statistical thinking involves using logic and common sense, not just mathematical calculations.

Populations and Samples

Definitions

• Population: The entire group you want information about.

• Sample: A smaller group selected from the population for study.

Census vs. Sample

• Census: Data from every member of the population.

• Sample: Data from only part of the population (often used due to cost or time constraints).

Types of Data

Quantitative Data (Numerical)

• Discrete Data: Countable values (e.g., number of students).

• Continuous Data: Measured values that can include decimals (e.g., height, weight, time).

Categorical Data (Qualitative)

• Non-numerical measurements (e.g., gender, eye color, survey responses).

Levels of Measurement

• Nominal: Categories only, no order (e.g., gender, eye color).

• Ordinal: Categories with order, but differences are not meaningful (e.g., letter grades).

• Interval: Ordered data, differences are meaningful, but no true zero (e.g., temperature in °C).

• Ratio: Ordered data, differences are meaningful, true zero exists (e.g., height, age).

Quick Level Summary: Nominal = categories; Ordinal = categories + order; Interval = differences, no zero; Ratio = differences + true zero.

Big Data

• Extremely large and complex data sets.

• Requires advanced software and data science techniques.

• Applications: Computer science, business, healthcare, etc.

Missing Data

Types of Missing Data

• Missing Completely at Random: Missingness is unrelated to any variable.

• Missing Not at Random: Missingness is related to the reason it is missing.

Handling Missing Data

• Delete cases: Remove records with missing values.

• Impute values: Replace missing values with estimated values.

Parameters vs. Statistics

• Parameter: Numerical value that describes a population.

• Statistic: Numerical value that describes a sample.

The Gold Standard in Experiments

• Random assignment to treatment and placebo groups is called the gold standard.

• Placebo: An inactive treatment (like a sugar pill) used to compare real effects vs. psychological effects.

Ways to Collect Data

• Observational Studies: Observe and measure without changing anything.

• Experiments: Apply a treatment and observe its effects.

Design of Experiments

• Replication: Use enough subjects to see real effects.

• Blinding: Subjects do not know their treatment group.

• Double-blind: Neither subjects nor researchers know group assignments.

• Randomization: Use chance to assign subjects to groups.

Sampling Methods

• Simple Random Sample: Every possible sample has an equal chance.

• Systematic Sampling: Choose every nth item.

• Convenience Sampling: Use data that is easy to obtain.

• Stratified Sampling: Divide into similar groups and sample from each.

• Cluster Sampling: Divide into clusters, randomly select clusters, and sample all members.

• Multistage Sampling: Combine multiple sampling methods in stages.

Types of Observational Studies

• Cross-sectional: Data collected at one point in time.

• Retrospective: Looks back at past data.

• Prospective: Follows groups into the future.

Confounding and Controlling Variables

• Confounding: Occurs when you cannot identify the true cause of an effect.

• Controlling Variables: Good experimental design helps reduce confounding.

• Completely Randomized Design: Randomly assign subjects to treatments.

• Randomized Block Design: Group similar subjects, then randomize within blocks.

• Matched Pairs Design: Match subjects in pairs and compare treatments.

• Rigorously Controlled Design: Carefully balance groups (very difficult).

Sampling Errors

• Sampling Error: Differences between sample result and true population value due to chance.

• Nonsampling Error: Human mistakes (biased questions, wrong data, etc.).

• Nonsampling Error: Errors from nonrandom methods like convenience sampling.

Statistical and Practical Significance

• Statistical Significance: Results are unlikely due to chance (usually means probability < 5%).

• Practical Significance: Asks if the result matters in real life, not just statistically.

Example: Losing 2.1 kg in a year may be statistically significant but not worth the effort for many people.

Common Pitfalls in Statistics

• Misleading conclusions

• Self-reported data

• Order of questions

• Nonresponse

• Missing data

• Misleading percentages

Key Takeaways (Exam Ready)

• Statistics is more than calculations; it requires critical thinking.

• Samples should represent the population.

• Voluntary response samples are biased.

• Statistical significance ≠ practical importance.

• Always question how data were collected.