Introduction of Biostatistics

Introduction to Biostatistics and Epidemiology

Statistics Overview: Branch of mathematics focusing on data collection, organization, presentation, analysis, and interpretation.
- Mathematical Statistics: Development of statistical inference methods; requires advanced mathematical knowledge.
- Applied Statistics: Application of mathematical methods across various fields (e.g., economics, psychology, public health).

Descriptive Statistics: Summarizes and presents data to simplify analysis.
Inferential Statistics: Generalizes findings from samples to populations, including estimations, hypothesis testing, and predictions.

Descriptive: "In 2011, there were 34 deaths from avian flu."
Inferential: "In 2025, the world population is predicted to be 8 billion."
More examples listed in the text demonstrate understanding of both statistics types.

Parametric Statistics: Assumes underlying normal data distribution; includes hypothesis testing based on population means.
- Assumptions: Normality, homoscedasticity, interval/ratio level measurements, absence of outliers.
Nonparametric Statistics: Does not assume any distribution; requires larger sample sizes to achieve equivalent power to parametric tests.

Definition: Study of how diseases and health conditions affect populations, aiming to understand and mitigate health issues.
- Historical Context: Originally focused on infectious diseases; now includes chronic diseases and health-related conditions.
Key Concepts:
- Distribution: Analysis of when, where, and in whom health issues occur.
- Determinants: Factors influencing health problems, including biological, social, and behavioral factors.

Variable: Characteristic that can vary (e.g., height, weight).
Data: Values represented by a variable; sets of measurements are called datasets.
Population: Complete set of subjects under study; Sample: Subset taken from the population.
Census: Data collection from a complete population; Parameter & Statistic: Measures from populations and samples respectively.

Levels of Measurement:
- Nominal: Categories without order (e.g., gender).
- Ordinal: Categories with rank but no precise differences (e.g., letter grades).
- Interval: Ordered categories with defined differences but no true zero (e.g., temperature).
- Ratio: Defined differences with true zero (e.g., weight).
Classification of variables by type, with examples.

Experimental Variables:
- Response Variable (dependent): Affected by other variables.
- Explanatory Variable (independent): Affects the response variable.
Correlation Studies: Predictor and criterion variables defined but correlation research does not classify them as independent or dependent.

Qualitative: Distinct categories (e.g., gender).
Quantitative:
- Discrete: Countable (e.g., number of children).
- Continuous: Measurable with infinite values (e.g., height).

Internal Data: Related to the organization’s activities.
External Data: Collected from outside the organization.
Statistical Data: Published figures by institutions; Nonstatistical Data: Information without quantitative representation.

Data Collection Methods: Observational, experimental; importance of both.
Types of Studies: Cross-sectional, retrospective, prospective outlined with definitions and examples.
Clinical Trials: Defined as experiments to evaluate interventions versus controls.

Various methods include questionnaires, interviews, registration, observation, and experiments, with an exploration of each.
Questionnaires: Closed-ended and open-ended types with characteristics outlined.
Interview Method: Various approaches discussed.
Observation and Registration Methods: Explained with examples.

Definitions and differences between target population and study population, with examples.
Criteria for Sampling Design: Representation, reliability, practicality, efficiency.
Probability Sampling Methods: Simple random, systematic, stratified, cluster, and multistage sampling detailed with procedures and examples.

Overview of diverse sampling techniques like convenience, purposive, and snowball sampling.
Quota Sampling: Types detailed, including proportional and non-proportional methods.
Additional examples provided for clarity.

Defined importance of estimating the proper sample size; includes methodologies such as census, replicating, tables, and formula applications.
Power Analysis: Importance in determining the sample size while considering effect size and significance levels.
G*Power Calculator: Explained with steps for conducting power analysis and its uses for various research designs.