Principles of Random Sampling, Variable Identification, and Bivariate Data Analysis

Fundamentals of Random Sampling and Bias Avoidance

  • Selection Criteria and Subjectivity:

    • Selecting a sample based on personal, non-random observations—such as choosing five individuals from a classroom because they are smiling or wearing a nice sweater—introduces subjective bias.

    • Visual or arbitrary picking fails to produce a true random sample because individual selection probabilities are not equal or objective.

  • Achieving Unbiased Selection:

    • Creating an unbiased, statistically valid random sample requires an objective selection process, such as utilizing a random variable or automated random selection tool, rather than discretionary choice.

Explanatory and Response Variables in Bivariate Data

  • Structural Setup of Bivariate Relationships:

    • Bivariate data explores the potential association between two variables plotted across two coordinate axes.

    • Explanatory Variable:

    • Plotted along the horizontal axis (xx\text{-axis}).

    • Represents the independent condition, cause, or predictor variable.

    • Response Variable:

    • Plotted along the vertical axis (yy\text{-axis}).

    • Represents the dependent outcome variable that responds to changes in the explanatory variable.

Case Study: Bill Amount vs. Tip Amount

  • Variable Assignment:

    • Explanatory Variable (xx\text{-axis}): Total bill amount.

    • Response Variable (yy\text{-axis}): Tip amount.

  • Underlying Logic:

    • The total bill serves as the explanatory variable because a customer first receives the bill and uses that specific amount as the basis to decide the tip.

  • Association and Direction:

    • Association Type: Positive association.

    • Relationship Trend: As the bill amount increases, the tip amount generally increases as well.

Discussion and Session Reflection

  • Group Discussion on Variable Classification:

    • Group inquiry focused on properly distinguishing between explanatory and response variables during graph analysis.

    • Specific outreach was made to group members, including Anne, to contribute perspectives on which factor acts as the explanatory variable.

    • The group reached a unanimous consensus confirming that the bill amount is the explanatory variable due to its causal influence on the tip amount.

  • Reflections on Session Dynamics:

    • Participants noted that the material and session structure were interesting.

    • Concerns were expressed regarding the intensity, pace, and long-term sustainability of the current session format.