Modul 1 - Konsep Dasar Statistika

Statistika vs. Statistik: Fundamental Definitions
📌 Statistika is defined as the overall process or methodology (the "how-to") of studying phenomena under uncertainty, involving data collection, analysis, interpretation, and conclusion drawing.
📊 Statistik refers specifically to the result or the numerical values derived from analyzing a sample data set (e.g., the result of a quick count).
🔍 The core difference lies in scope: Statistika is the method, while Statistik is the outcome/value, often derived from a sample.

Data Collection and Results Terminology
🏛️ When using the entire population data set, the method is called a Sensus (Census), and the result is termed a Parameter.
🔬 When using a sample data set, the method is called a Survei (Survey), and the result is termed Statistik.
🔮 Statistics derived from samples are inherently uncertain because they aim to predict the true population Parameter before the final census results are available.

Classification of Statistical Analysis
📝 Statistika Deskriptif (Descriptive Statistics) is used for describing and summarizing collected data, covering measures of central tendency (mean, median, mode) and dispersion (variance, standard deviation).
🔮 Statistika Inferensial (Inferential Statistics) is used for drawing conclusions and making inferences or decisions (like hypothesis testing) about the population based on sample data.
⚖️ The decision to use Parametric vs. Non-Parametric statistics depends on data characteristics, including whether the data is quantitative/qualitative, the sample size ($n>30$ generally favors parametric), and whether the data follows a Normal Distribution.

Data Scales and Parametric/Non-Parametric Choice
🏷️ Kualitatif Data (Categorical) is divided into Nominal (no inherent order, e.g., gender) and Ordinal (has a specific order, e.g., education level: SD, SMP, SMA). These typically require non-parametric statistics.
🔢 Kuantitatif Data (Numerical) is divided into Interval (equal distances between points, but no true zero, e.g., temperature) and Rasio (true zero exists, allowing all mathematical operations, e.g., weight, height). These levels are typically analyzed using parametric statistics if assumptions are met.
📏 A key prerequisite for using Parametric Statistics is assessing if the sample characteristics align with the population characteristics, often checked using the Normal Distribution assumption.

Key Points & Insights
➡️ Understand that statistical conclusions are often characterized by ketidakpastian (uncertainty), not incorrectness, because they are based on sample data.
➡️ Descriptive Statistics explains *what is* (e.g., Class A GPA is 3.5), while Inferential Statistics answers *if there is a difference* or *why* (e.g., Is 3.5 significantly different from Class B's 3.0?).
➡️ When data is non-normal (mean, median, mode are not symmetric), one must default to using Statistika Non-Parametrik for robust conclusions.

📸 Video summarized with SummaryTube.com on Jan 22, 2026, 00:44 UTC