Course Schedule: Weeks 1 and 2 Topics

Course Schedule and Topic Overview

• Week 1 ($\text{Aug 24} - \text{Aug 28}$):

  • Primary Topics Covered: Sampling and data, types of variables, levels of measurement; parameter, population, statistics, sample, etc.

• Week 2 ($\text{Aug 31} - \text{Sept 4}$):

  • Primary Topics Covered: Observational studies and Experiments, Experimental design and ethics.

Week 1: Core Statistical Concepts – Population, Sample, Parameter, and Statistic

• Population:

  • Definition: The entire collection of all individuals, objects, items, or measurements whose properties are to be analyzed and about which inferences are to be drawn.

  • Scope: The population represents the grand totality of the target group under investigation.

• Sample:

  • Definition: A specific subset or sub-collection of elements drawn from a population.

  • Purpose: Because analyzing an entire population is frequently unfeasible, costly, or time-consuming, samples are gathered to collect data and make statistical inferences regarding the population as a whole.

• Parameter:

  • Definition: A fixed numerical measurement describing a characteristic of a population.

  • Characteristics: Parameters are theoretical or true values of a population and are typically denoted by Greek letters (e.g., population mean $\mu$, population standard deviation $\sigma$, population proportion $p$).

• Statistic:

  • Definition: A numerical measurement calculated from sample data that describes a characteristic of the sample.

  • Characteristics: Statistics vary from sample to sample due to sampling variability and are denoted by Roman letters (e.g., sample mean $\bar{x}$, sample standard deviation $s$, sample proportion $\hat{p}$).

• Data:

  • Definition: Raw observations, measurements, counts, responses, or facts collected for analysis.

• Sampling and Data Collection:

  • The systematic methodology used to select a representative sample from a target population to ensure that findings can be validly generalized.

Week 1: Types of Variables

• Definition of Variable:

  • A characteristic, attribute, or property of interest that can assume different values for different individuals or objects in a population or sample.

• Qualitative (Categorical) Variables:

  • Definition: Variables that describe attributes, qualities, categories, or labels rather than numerical quantities.

  • Examples: Gender, hair color, blood type, vehicle make, zip code, preference ratings.

• Quantitative (Numerical) Variables:

  • Definition: Variables that represent counts or measurements expressed numerically, allowing for meaningful mathematical operations such as addition or averaging.

  • Discrete Variables:

  • Definition: Quantitative variables whose values are countable, finite, or countably infinite, typically resulting from counting process instances.

  • Examples: Number of students in a class, number of defective items in a shipment, number of cars passing an intersection.

  • Continuous Variables:

  • Definition: Quantitative variables whose values are uncountably infinite and can take on any real value within a given continuous interval, typically resulting from measurement process instances.

  • Examples: Height, weight, temperature, exact time to complete a test.

Week 1: Levels of Measurement

• Nominal Level of Measurement:

  • Definition: Data consists of names, labels, categories, or qualities only.

  • Properties: Data cannot be arranged in an ordering scheme or ranked in a meaningful mathematical order. Arithmetic operations cannot be performed.

  • Examples: Eye color, marital status, major field of study, gender.

• Ordinal Level of Measurement:

  • Definition: Data consists of values that can be arranged in a specific order or rank.

  • Properties: Differences between data values either cannot be determined or are mathematically meaningless, even though relative positions or rankings are established.

  • Examples: Class rankings (first, second, third), letter grades ($\text{A}$, $\text{B}$, $\text{C}$, $\text{D}$, $\text{F}$), survey satisfaction levels (poor, fair, good, excellent).

• Interval Level of Measurement:

  • Definition: Quantitative data where values can be ordered and meaningful, precise differences between data values can be calculated.

  • Properties: There is no natural or absolute zero starting point where zero signifies the complete absence of the quantity. Ratios of values are meaningless.

  • Examples: Temperature in degrees Fahrenheit or Celsius (where $0^\circ\text{C}$ does not mean no temperature), calendar years ($\text{Year } 1000$, $\text{Year } 2000$).

• Ratio Level of Measurement:

  • Definition: Quantitative data with all the properties of the interval level, augmented by a true, natural zero point indicating complete absence of the quantity.

  • Properties: Differences and ratios between data values are meaningful; a value of $20$ represents twice as much as a value of $10$.

  • Examples: Height, weight, distance, elapsed time, financial balance.

Week 2: Observational Studies vs. Experiments

• Observational Studies:

  • Definition: Research studies in which investigators observe, measure, and record specific characteristics or outcomes without attempting to modify, manipulate, or intervene in the subjects or environment.

  • Characteristics: The researcher acts as a passive recorder. Cause-and-effect relationships cannot be firmly established due to potential confounding variables.

  • Types:

  • Cross-sectional studies (data collected at a single point in time).

  • Retrospective studies (data collected from past records or history).

  • Prospective studies (cohorts followed over time into the future).

• Experiments:

  • Definition: Studies in which researchers deliberately apply a treatment, condition, or stimulus to subjects (experimental units) to observe and measure the resulting responses or effects.

  • Characteristics: Allows researchers to control variables and establish direct cause-and-effect relationships between treatments and outcomes.

Week 2: Experimental Design Principles and Ethics

• Key Principles of Experimental Design:

  • Control Group and Treatment Group:

  • Experimental units are separated into groups: a treatment group receiving the active intervention and a control group receiving no treatment or a placebo.

  • Randomization:

  • The process of randomly assigning subjects to different treatment or control groups to minimize bias and balance out unexpected confounding factors.

  • Replication:

  • Repeating an experiment on a sufficiently large sample size of experimental units to confirm validity and ensure statistical reliability.

  • Blinding and Double-Blinding:

  • Single-Blinding: Subjects do not know whether they are receiving the real treatment or a placebo.

  • Double-Blinding: Neither the subjects nor the researchers assessing outcomes know group assignments, eliminating both participant expectations and researcher evaluation bias.

  • Placebo Effect:

  • A psychological or physical improvement observed in control group subjects who believe they are receiving an active treatment when receiving an inactive substance (placebo).

  • Confounding Variables:

  • Uncontrolled or extraneous variables that influence the outcome variable, making it difficult to isolate the true effect of the treatment variable.

• Ethics in Experimental Design and Research:

  • Informed Consent:

  • Subjects must be fully educated regarding the nature, risks, potential harms, benefits, and procedures of the study before voluntarily agreeing to participate.

  • Institutional Review Board (IRB) / Ethics Committees:

  • Research protocols involving human or animal subjects must undergo formal ethics review and approval to ensure welfare and safety protection.

  • Avoidance of Harm and Risk Minimization:

  • Experiments must strictly adhere to safety standards, weighing potential risks against anticipated scientific or societal benefits.

  • Confidentiality and Anonymity:

  • Personal identifying details of participants must be protected and maintained securely throughout data collection, analysis, and publication.