Lesson 2: Overview of Data
1. What is Data?
• Definition: Raw, unorganized information that can be recorded, stored, and processed.
• It can consist of:
o Facts: e.g., "Temperature is 30°C".
o Figures: e.g., sales numbers, exam scores.
o Observations: e.g., a customer buying a product.
o Symbols: e.g., codes, text, images.
• Characteristics:
o Raw form: Data by itself may not have meaning until processed or analyzed.
o Various formats: Numbers, text, audio, video, images, sensor readings, etc.
o Source of information: Once processed, data becomes information that supports decision-making.
2. Key Concepts:
• Variable: A characteristic or quantity of interest that can take on different values.
• Observation: A set of values corresponding to a set of variables.
• Variation: The difference in a variable measured over observations.
• Random/Uncertain Variable: A quantity whose values are not known with certainty.
3. Data Flow:
• DATA → processed into → INFORMATION → interpreted into → KNOWLEDGE
• Example:
o Raw data: 120, 130, 125, 140
o Information (after processing): "The average sales this week were 128 units."
o Knowledge (after interpreting): "Since the sales is consistent and within the margins, we can
continue selling at the same price."
4. Types of Data:
• Structured Data:
o Organized in rows and columns (like databases).
o Easy to store, search, and analyze.
o Example: Customer names, transaction dates, sales amounts.
• Unstructured Data:
o No predefined format.
o Difficult to process without specialized tools.
o Example: Emails, social media posts, images, videos.
• Semi-Structured Data:
o Has some organizational structure but isn't fully tabular.
o Example: JSON, XML, log files.
5. Sample and Population:
• Population: All elements of interest.
• Sample: A subset of the population used for study.
6. Qualitative vs. Quantitative Data:
• Qualitative Data:
o Definition: Descriptive, non-numeric information about qualities or characteristics.
o Nature: Subjective.
o Form: Words, symbols, images.
o Examples: Customer feedback, interview responses, colors, emotions.
o Purpose: To understand meanings, concepts, and experiences.
o Collection Methods: Interviews, focus groups, observations, open-ended surveys.
o Analysis: Thematic analysis, coding, pattern recognition.
o Output: Descriptive reports, themes, categories.
• Quantitative Data:
o Definition: Numeric information that can be measured or counted.
o Nature: Objective.
o Form: Numbers, percentages, statistics.
o Examples: Height, weight, temperature, sales figures.
o Purpose: To quantify variables and analyze statistical relationships.
o Collection Methods: Experiments, measurements, closed-ended surveys, structured
observations.
o Analysis: Statistical analysis, mathematical models.
o Output: Charts, graphs, averages, correlations.
7. Levels of Measurement:
• Classify data based on how it can be measured and analyzed, helping determine appropriate statistical
methods.
• Nominal Level (Lowest):
o Definition: Data classified into distinct categories with no order.
o Key Points: Labels or names only, no ranking possible.
o Examples: Gender, blood type, types of cuisine.
o Common Statistical Tests: Mode, Chi-square.
• Ordinal Level:
o Definition: Data arranged in a meaningful order, but differences between ranks are not equal.
o Key Points: Shows relative position, exact difference unknown.
o Examples: Customer satisfaction ratings (1-5 stars), education level, race rankings.
o Common Statistical Tests: Median, Spearman correlation.
• Interval Level:
o Definition: Numerical data with equal intervals between values, but no true zero.
o Key Points: Differences are meaningful, zero does not mean "absence".
o Examples: Temperature (°C or °F), dates (years), IQ scores.
o Common Statistical Tests: Mean, Pearson correlation, t-test, ANOVA.
• Ratio Level (Highest):
o Definition: Numerical data with equal intervals and a true zero point.
o Key Points: All arithmetic operations possible, ratios are meaningful.
o Examples: Height, weight, age, distance, income.
o Common Statistical Tests: Same as interval (Mean, Pearson correlation, t-test, ANOVA).