SM

Research Issues

Definition (#f7aeae)

Important (#edcae9)

Extra (#fffe9d)

Variables:

  • Any event, situation, behavior or individual characteristic.

  • Each variable is a general class, there are specifics within that can vary called levels.

  • Some levels contain values.

Types:

  1. Situational variable: Related to the characteristics of a situation. Ex: environmental factors.

  2. Response variable: Responses we receive from participants. Ex: problem solving, recall time.

  3. Participant variable: Individual differences. Ex: gender, educational background.

  4. Mediating variable: Variable which explains the relation between the IV and DV.

    • Explains why there is a relation between the 2 variables.

    • A potential mechanism for an IV to produce changes in the DV.

    • Ex: Socio-economic status and reading ability in children. Parental educator level can be a mediator.

  5. Moderating variable: Variable that affects the strength of the relation between the IV and DV.

    1. Ex: Link between video game playing and aggression.

    2. Moderator: Age, kids may be more aggressive but not for adults.

  1. Independent variable: The variable researchers can change or control.

  2. Dependent variable: The variable being measured or observed based on that change.

Variables are abstract concepts, we need to transform them into concrete form of manipulations.

4 Categories: Based on the measurement of the variable.

  1. Nominal (categorical): Color, gender, nationality.

  2. Ordinal: SES, educational level, ability level.

  3. Interval: No true zero. Ex: Temperature, credit score.

  4. Ratio: Consists of true zero. Ex: height, weight.

Operational Variables:

  • Variables are abstract concepts, to transform them into concrete form of manipulations.

  • These variables need to be defined in terms of specific methods used to measure or manipulate it. This allows for empirical research.

Relationship between Variables:

  1. Positive correlation:

    • As 1 variable increases, there’s an increase in the 2nd variable.

    • Points lie in a straight line in a positive gradient.

  2. Negative correlation:

    • As 1 variable increases, there’s a decrease in the 2nd variable.

    • Points lie in a straight line in a negative gradient.

  3. No correlation:

    1. There is not connection between the variables.

    2. There is no pattern in the points.

Patterns between Variables (Graphs):

  1. Positive Linear Relationship: Increase in value of 1 variable, is accompanied by increases in the values of the second variable.

  2. Negative Linear Relationship: Increase in the value of 1 variable is accompanied by decreases in the value of the second variable.

  3. Curvilinear Relationship: Increases the values of one variable, are accompanied by both increases and decreases in the values of the other variable.

  4. No Relationship: There’s no relationship, the graph is just a flat line.

Correlation coefficient: A numerical index of the strength of relationship between variables.

Error Variance:

  • When measuring a variable (test scores, reaction times), not all differences between people are due to the variable we’re studying.

  • Some variability comes from random factors we can’t control; this is called random or error variance.

  • The goal of scientific research is to reduce random variability by identifying systematic relationships between variables.

Determining whether variables are related:

  1. Non-Experimental Method:

    • Observing variables of interest.

    • Asking people to describe their behavior, directly observing behavior.

    • Difficult to determine which variable causes the other.

      • 1st problem: Directionality.

      • 2nd problem: Extraneous variables; Danger that no direct causal relationship exists between the two variables.

  2. Experimental Method:

    • Direct manipulation and control of variables.

    • The researcher manipulates the first variable of interest and then observes the response.

    • Able to reduce ambiguity better, and uncertainty in making conclusions.

      • Experimental control: All extraneous variables are constant so they cannot influence results, ensuring any differences between groups are due to the manipulated variable.

      • Randomisation: Randomly assigning participants to groups makes it equally likely that any extraneous factors (age, prior knowledge) are spread out across conditions.

Issues with Experimental Research:

  1. Experiments are artificial:

    • High degree of control in lab experiments can limit the generalizability of the results.

    • Researcher can sometimes do field experiments, where the variable is manipulated in a natural environment.

  2. Ethical and Practical Considerations:

    • Child rearing practices, alcoholism, social issues, maternal employment.

  3. Participant Variables:

    • Gender, ethnicity, nationality, marital status cannot be allocated randomly.