L1 MATH1005
Data Collection
Course: Quantitative Methods (MATH1005)
Chapter: 1
Instructor: Qianru Shang (Jennifer)
Contact: gianru.shang@tudublin.ie
Key Definitions
Elements: Entities on which values are collected.
Variable: Characteristic that can vary (e.g., sales, expenses).
Data: Different values associated with variables.
Population: Entire collection of subjects under consideration.
Sample: Portion of the population selected for analysis.
Parameter: Numerical measure for a population characteristic.
Statistic: Numerical measure for a sample characteristic.
Population and Sample
Use parameters to summarize population features.
Use statistics to summarize sample features.
Inference about the population is drawn from sample data.
Data Sources
Primary Data: Collected directly by the user for analysis.
Secondary Data: Collected by someone else and used for analysis.
Primary Collection Methods:
Experiment
Observation
Survey
Secondary Compilation Sources: Data from organizations or individuals.
Types of Data
Categorical Data: Can be placed into categories (e.g., gender).
Numerical Data: Represents quantities. Further divided into:
Discrete Data: Countable values (e.g., number of subscriptions).
Continuous Data: Any value within a range (e.g., weight).
Cross-Sectional Data: Collected at a specific point in time.
Time Series Data: Collected across several time periods.
Scales of Measurement
Nominal: Data categorized by names/labels (e.g., brands).
Ordinal: Data with meaningful order (e.g., ratings).
Interval: Data with meaningful differences but no true zero (e.g., temperature).
Ratio: Data with meaningful ratios and a true zero (e.g., weight).
Census vs Sample
Census: Study of every unit in a population (complete enumeration).
Sample: Subset representing the population.
Pros and Cons of Census
Pros: Accurate, no sampling error, detailed info on subgroups.
Cons: Time-consuming, expensive, may not reach all units.
Pros and Cons of Sample
Pros: Cost-effective, quicker results, can be representative.
Cons: Possible non-representativeness, subject to sampling error.
Sampling Methods
Non-Probability Sampling: Selection without known probabilities.
Convenience Sampling: Based on easy access.
Judgment Sampling: Based on expert opinion.
Probability Sampling: Selection based on known probabilities.
Simple Random Sampling: Every individual has an equal chance of selection.
Systematic Sampling: Selection at regular intervals.
Stratified Sampling: Dividing population into groups and sampling each.
Cluster Sampling: Dividing population into clusters and selecting clusters for analysis.
Advantages and Disadvantages of Sampling Methods
Simple Random & Systematic: Simple but may not represent population well.
Stratified: Ensures representation.
Cluster: Cost-effective but less efficient.
Summary
Key definitions and concepts introduced, including data collection sources, types of data, census vs sample, and sampling methods.