Unit 1.2: Research Methods and Critical Thinking with Psychology

Unit 1.2: Research Methods and Critical Thinking with Psychological Science

Hurdles to Research

  1. Overconfidence:
  • The number one hurdle to conducting research is getting over your overconfidence.
    • The Overconfidence Test: Everyone is really overconfident. I got 9/14. Not one person got a 100%, even though it was extremely easy to.
  1. Hindsight Bias
  • Tendency to believe, after learning the outcome of something, that you knew it all along. “Hindsight is always 20/20.”
    • Example: Elections. People are like “yeah I knew ____ was going to get elected” even though it was unexpected.
  1. Hawthorne Effect
  • Participants being told they are being observed can alter results of an experiment.
    • Named after the experiment in the Hawthorne Electric Plant.
  1. The Barnum Effect
  • The tendency for people to accept very vague or general characterizations of themselves and take them to be accurate.
    • Example: Zodiac Signs !!

The Scientific Method

  • The Scientific Method is a self correcting process for evaluating ideas with observation and analysis. It puts ideas into an objective pass/fail test.
    • Theory: explains behaviors or events by offering an idea
    • Hypothesis: a testable prediction implied by a theory
    • Empirical Investigation: testing the hypothesis by collecting objective information through direct experiences.
    • Operational Definitions
    • Explain what you mean in your hypothesis
    • How will the variables be measured in real life
    • How you will operationalize the variables, which tells if the study is valid and reliable.
    • Replication: Repeating a research study to see if the basic finding can be reproduced

Types of Research

  1. Case Study
  • A detailed study of one or a few subjects
  • Descriptive research, so it can’t be used to draw correlations
  1. Naturalistic Observation
  • Watching subjects in their natural environment
  • Avoids the Hawthorn Effect because the environment isn’t being manipulated.
  • Cannot be used to show cause and effect but the cause cannot be manipulated by the researchers.
  1. Survey Method
  • Most common type of study, can measure correlation. Typically has a low response rate.
  • Cheap and fast, but need a good random sample.
  • Random Sampling
  • You need to identify the population you want to study.
  • A sample is a representative of the population that you want to study.
  • A random sample allows all students to have an equal chance of participating, which prevents sampling bias.
  1. Correlational Method
  • Correlation expresses a relationships between two variables.
  • Does not confirm causation.
  • r = correlational coefficient
  • Ranges from -1 to +1, higher correlation is further from 0
  • Positive correlation is when variables go in the same direction
  • Negative correlation is when variables go in opposite directions
  1. Experimental Method
  • Looking to prove causal relationships (cause and effect)
  • Directly manipulating variables in order to come to a conclusion.
    • Example: abilify has been proven to reduce schizophrenic symptoms
  • Confounding variables are variables that could cause change in an experimental group that is not the independent variable being studied in an experiment.
    • These can cause illusory correlations, perceived correlations where they don’t exist.
  • Random Assignment is randomly assigning groups in an experiment controls for confounding variables.
    • Vs. a random sample: Random assignment is randomly assigning groups in an experiment, whereas random sample is randomly choosing subjects to participate in an experiment. Both of these methods allow for representative results.
  • Experimental group is the group exposed to the treatment
  • Control group is not exposed to treatment and is used as a reference to show effect of the treatment.
  • Double Blind Procedure is when both the subjects and experimenters don’t know what they have been assigned (real vs. placebo). This prevents confounding variables.
  • Placebo effect is a result caused by the subject’s expected results.
  • Independent vs. Dependent Variable: Independent is the thing being tested, manipulated by experimenter. Dependent is the thing, or result, being measured.

Statistics in Psychology

Descriptive Statistics

  • Describes sets of data, can use a frequency distribution using polygons or histograms.
  • Uses Distribution
  • Terms to describe distribution: Mean, Median, Mode, Range, Standard Deviation, Bell Curve, 68-95-99 rule, Outliers, Skews.
  • Scientists are looking for small standard deviation when experimenting because it implies more consistency.
    • Right skew == Positive skew
    • Left skew == Negative skew

Influential Statistics

  • Numerical data that allows for generalization/inference from the sample of something being true for a population.

Statistical significance: a statistical statement of how likely it is that an obtained result occurred by chance. Data has statistical significance if the difference between two groups is large enough to conclude cause and effect.