Notes on Independent and Dependent Variables from Transcript

Independent and Dependent Variables

  • There are always two basic types of variables in hypotheses: dependent and independent.
  • There is usually only one dependent variable per hypothesis — you're trying to figure out one outcome at a time. This helps avoid confusion when building and testing hypotheses.
  • There can be more than one independent variable in a hypothesis (i.e., multiple factors that might cause change).
  • A hypothesis links an independent variable to a dependent variable (the thing you expect to change).
  • The dependent variable is what you measure or what you expect to change as a result of the independent variable(s).
  • If a statement would imply more than one dependent outcome, you should split it into separate hypotheses for clarity (e.g., graduation and getting a full-time job can be separate dependent outcomes).
  • Examples illustrate how a single independent variable can influence different dependent variables in different studies.
  • The direction of the relationship is not fixed: with some variables, higher values may lead to higher outcomes (up-up), with others, higher values may lead to lower outcomes (down-down), and in some cases the direction can vary depending on context.
  • In practice, you should identify frequency and specific aspects of the independent variable (e.g., how often, how much) as part of the independent variable.

Hypotheses Structure and Variable Roles

  • Independent variable: the variable thought to cause change in the dependent variable; it’s the cause or predictor.
  • Dependent variable: the variable thought to be affected by the independent variable; it’s the outcome.
  • There can be more than one independent variable in a hypothesis; you can test multiple potential causes at once.
  • Example scaffold: "If I work out every day, [outcome] will happen" where
    • Independent variable: frequency of working out (e.g., every day)
    • Dependent variable: the outcome (e.g., blood pressure).
  • Example: "Students whose parents have a college degree are more likely to graduate from college".
    • Independent variable: parents' education level (college degree or not).
    • Dependent variable: graduating from college.
  • This can be followed by a second study with the same independent variable but a different dependent variable (e.g., getting a full-time job).
  • Note that the independent variable may influence multiple outcomes; these should be treated as separate hypotheses for clarity.
  • The process is not necessarily linear or monotonic; relationships can vary across outcomes and contexts.

Examples Discussed: Variables in Action

  • Example 1: Working out every day → blood pressure change
    • Independent variable: frequency of working out (every day)
    • Dependent variable: blood pressure
    • Concept: if workouts go up, blood pressure should go down (direction may be negative, but not guaranteed in all contexts).
    • Possible formal relation: rac{d\text{blood_pressure}}{d\text{workout_frequency}} < 0 (in many cases).
  • Example 2: Parental education → college graduation
    • Independent variable: parents' education level (e.g., college degree or not)
    • Dependent variable: graduating from college
  • Example 3: Parental education → full-time job
    • Independent variable: parents' education level
    • Dependent variable: obtaining a full-time job
  • Example 4: Multiple independent variables influencing graduation outcomes
    • Potential independent variables include:
    • Study habits (e.g., study duration per week)
    • Sleep (e.g., hours of sleep per night, seven to eight hours)
    • Living proximity to the university
    • Amount of study time (e.g., 10 hours per week)
    • Extracurriculars or sports
    • Having a job while studying
    • These variables could influence the likelihood of graduating, hence they can be tested as independent variables.
  • Example 5: Women with mothers who work full-time
    • This example flips the direction of cause and effect to test a different dependent variable:
    • Independent variable: mother’s work status (full-time vs not)
    • Dependent variable: whether the daughter/woman works full-time
    • Other independent variables to consider: one-parent household status, salary/pay considerations, length of time at a current job, housing status (owning a house), having children, etc.
    • Note: Salary can be related to employment outcomes but may overlap conceptually with pay; treat as related but distinct variables if measuring different constructs.
    • Additional factors that could influence getting a full-time job: longer duration at a job, benefits (health, dental), etc.

Independent and Dependent Variables: Definitions and Roles

  • Independent variable: the variable thought to cause change in another variable; the cause/predictor.
  • Dependent variable: the variable expected to change as a result of the independent variable(s); the outcome.
  • There can be more than one independent variable in a study, and there is usually one dependent variable per hypothesis.
  • When a study could imply more than one dependent outcome, split into separate hypotheses for clarity.
  • Relationships can be monotonic or non-monotonic; direction can vary across contexts and outcomes.

The Process and Classroom Context

  • The material emphasizes that social science is a process of learning; not everything clicks immediately.
  • This is a practical, exercise-driven approach rather than a full statistics class.
  • The instructor encourages brainstorming multiple independent variables and practicing how to structure hypotheses.
  • Ethnic Studies and related foundational courses are mentioned as part of the broader educational context.
  • The instructor plans to run more examples later, acknowledging that more practice helps solidify understanding.

Quick Reference: Notation and Simple Formulas

  • General form of a hypothesis:
    • Independent variable X,
    • Dependent variable Y,
    • Hypothesis: If X, then Y (direction of effect may be positive or negative).
  • Functional form (conceptual): Y = f(X)
  • If you want to indicate direction,
    • Positive relationship: \frac{dY}{dX} > 0
    • Negative relationship: \frac{dY}{dX} < 0
  • For probability outcomes (e.g., likelihood of graduating): P( ext{graduate}) = f( ext{parents_education}, \text{other factors})

Practical Takeaways for Study Design

  • If more than one outcome is of interest, split into separate hypotheses with their own dependent variables.
  • Start with a core independent variable (e.g., parental education) and brainstorm other factors that could influence the dependent variable (e.g., study habits, sleep, housing).
  • Consider both direct effects and potential confounding or interacting factors (e.g., employment during school, hours worked, proximity to school).
  • Be mindful of overlapping constructs (e.g., salary vs. pay); define clearly what each variable is intended to measure.
  • Use this framework to design simple, testable hypotheses and prepare for more complex analyses later in the course.