Why do we say that knowledge is flawed?
We are constantly striving for accuracy and there are several interference’s which cause us to make mistakes. (Think about who does the research - can cause bias)
Explain the two types of broad inquiry.
Deductive methods = Go to the library, read ideas, do the next logical step.
Inductive methods = Start by first immersing yourself in data, do a lot of observations, begin to see patterns, for a hypothesis, map out ideas, ask others later for verification as a way of avoiding bias.
Theory
A system of interrelated ideas that is used to explain a set of observations.
A good theory is testable - we are able to gather evidence for or against.
Research Hypothesis
A tentative statement about the relationship between 2 or more variables. (We may not always do one of these, depending on the type of study) It is frequently formulated from a theory.
Operational definitions
How do we define what we are studying. (Describe the actions or operations that will be made to study/test the variables)
Example - Measuring intelligence
Naturalistic Observation
it is an empirical design, not experimental. (It has gotten a lot easier to do this undetected due to advances in technology)
You watch without interfering. (People can and do change their behaviour when they know that they are being recorded)
Involves careful systematic recording.
Analyze - Usually this step occurs after as it is difficult to do this as new events are unfolding.
It is descriptive. - The focus is on describing the data.
Pro: They take place in natural environments and provide rich data. (They have ecological validity - they are more accurate)
Con: It is hard to be un-obtrusive – observer effect, you cannot conclude causality on the basis of your study design.
You can still make causal arguments on the basis of your theory but you are not doing it on the basis of your study design.
Empirical
A systematic study
Correlational Desgin
The study is done to determine whether two or more variables are related. If so, how are they related? (Positively/swing, negatively/teeter-totter?) Correlation is NOT the same as causation.
What is the third variable problem?
The possibility that a third variable causes the changes in the two variables that were studied.
(Video game, aggression, and lack of parent supervision example)
Third Variable Correlation
The fact that two variables are correlated only because each is causally related to the third variable.
Ginsburg and Miller study example on children at the zoo. (The importance of operational definitions)
The researchers went and watched boys and girls at the zoo. They were studying whether boys or girls displayed more risky behaviour. Their definition of risky behaviour was: riding elephants, petting animals, feeding animals, climbing steep embankments. After observing, the concluded that boys engaged in more risk taking behaviours. But, were they really measuring what they said they were?
Case study
A study of an individual person. Usually these are done on people with unique neurological impairment. Pro: Yields rich, contextual information. It allows us to study things that we otherwise could not.
Con: Limited generalizability, hard to argue for causal explanations; causal explanations cannot be argued for on the basis of study design alone.
Surveys
This is a common study design. They involve questionnaires that can ask specific questions.Pro: Large data sets, cheap, some issues are better suited to be collected anonymously (sexual habits)
Con: Accuracy (subjects will tell people what they think you want to hear), question bias, there may be low response rates (those who respond are not representative of the population). There are several different types of sampling bias – non-response, voluntary, convenience
Experiments
One of the most popular forms of study designs and for good reason, it is quite powerful. It allows us to strip away most variables and focus on very specific things.
Example: Does colour influence our perceptions as black and white are associated with evil and good in western culture?
Variable
Measurable condition, event, characteristics, behaviours, that can take on two or more levels.
Independent Variable
This is the variable you manipulate and set up. Does this variable have an effect on the dependent variable?
Dependent variable
Variable that you hope to observe change in
Experimental condition
A condition that exposes the subject to a version of the independent variable. It must have at least two conditions. This is also called levels of the independent variable.
Control Condition
Condition identical to the experimental conditions except it has a different value.
Random assignment
Placing all people into conditions by chance. In psychology statistics it is required that there is at least one random element as it helps to spread out the individual differences. It also helps to make sure that there isn’t a third variable influencing the results.
Confounding variable
Variables that are linked to levels of the independent variable. They can offer an alternative explanation for your findings. (Fever medication example with uni students and seniors)
Random sampling
Using a random process to choose your participants.
Quasi Experiments
Experiments that aren’t true experiments because the two groups that are tested already existed.
What are some of the pros and cons of experiments?
Pros: We can infer causation on the basis of study design (argue that IV caused DV because we set it up that way)
Cons: They artificial and may lack external validity, conditions may be set up to support hypothesis
