Chapter 2- The ways and means of Psychology
Psychologists try to use a scientific method that can be applied to a general situation (as in, it’s not too
specific, but it’s still precise for the event).
Scientific method= an approach in which data is collected through observation/ experiments under a
guided set of rules. Focuses on cause- and-effect, mostly associated with experiments rather than
Two methods with the least sets of rules:
Naturalistic observation= the study of animals in their natural environment (biological +social science,
ex: Charles Darwin, Montessori).
Clinical observation= the study of animals in psychological diagnosis or treatment.
Correlational Study= A method in which an observation is linked to measurements, like environmental
events, physical/social characteristics, etc.
Experiment= A scientific method in which the experimenter has the most control in setting and can thus
see results based on events that they made happen.
EX: Figure 2.1= some people can see SIRD stereogram in 3D faster while
others take longer or can’t see it at all. Why? Naturalistic observation can
indicate this observation and maybe give some context about it.
Experiments, however, can bring evidence to those assumptions that you
1) Identify the problem. Make a hypothesis on the cause-and-effect among the variables. Ex: If you
expect to see an image, then you’re more likely to see that image in the SIRD stereogram.
2) Design the experiment. Manipulate ONLY the independent variable, and see how it affects the
dependent variable. Ex: Producing an expectation (independent) can lead to different sorts of
3) Perform the experiment. Get the material, recruit researchers and participants. Sort the
participants into a control group and an experimental group. Perform experiment, observe.
4) Examine data and evaluate the hypothesis. Use statistical analysis to confirm or reject the
5) Communicate the result. Write out your experiment in an article and send it to journal
publishers for review. Present findings in conferences and conventions. Hypothesis: A statement that suggests that there is a cause-and-effect relationship between two or
Theory: An elaborate form of a hypothesis that uses related hypotheses and facts to cover a larger
aspect of nature. They should lead to testable hypotheses that can be proven wrong or supported.
Theories and hypotheses are started off with observation, so that the scientist has more knowledge on
the subject. Observers try to be the least disruptive when they’re in the subjects environment (think of
naturalists watching an animal in their natural habitat). Meanwhile, in clinical observation, the observer
does interact with the patient but records observations on the patient’s behavior in a more detailed
Survey study: Method of observation where you interact with participants to gain answers to
Variable: something that can change its value. Ex. Temperature, expectation, absence/presence of
Manipulate: to ‘handle’ independent variables and see their correlations with dependent ones.
Experimental group: A group exposed to the manipulated independent variable. Ex: An experience was
induced by the experimenter that would make the group expect to see the SIRD stereogram.
Control group: A group that is exposed to the independent variable with no value. Ex. This group does
not have any expectation to see the SIRD stereogram.
Independent variable: the variable that is manipulated in the experiment. It induces different values to
the dependent variable in a cause-and-effect event.
Dependent variable: the variable that is measured in order to see how much it depends on the
Nominal fallacy: Where an event is falsely identified and labeled. Ex: We say someone is angry based on
seeing their behavior, but anger is an internal state that might not necessarily cause that behavior,
unless we know for sure. You have to identify previous internal and external events that could have
caused that behavior to happen. You also have to be careful whether the previous events are related to
the behavior. (ex. You exit the train because of an announcement, not because someone picked up a
Operational definition: In an experiment, you will decide what the variables will be and you define them
based on what the operations are in your experiments. However, you have to be careful how you set up
these definitions, else they lead to different results among research experiments.
Validity: How accurate a defined variable is for testing an experiment. Confounding of variables: Extra independent, and unwanted, variables are introduced to the
experiment and thus interfere with the dependent variable. When this occurs, it is harder to distinguish
what lead to the state of the dependent variable, thus is hard to make a conclusion.
Habituation: When in an experiment, you expose the subject to a stimulus repeatedly, they will
eventually get used to it and the response will weaken. Ex: the ‘weakened predator’ experiment, where
the experiment was flawed, because they exposed a bird to shapes resembling predators in an order of
most similar looking to predators to least (as in, the shape of a triangle) repeatedly in the exact order.
The solution to this would be counterbalancing: the subject is exposed to a stimuli in a random order,
and the results are recorded using averages.
Reliability: the measurement of a variable that leads to consistent results when you measure it
repeatedly. Ex: measuring weight using a scale is reliable.
When running an experiment, it is often built objectively so that other scientists can repeat the
experiment and can get the same results. However, in many cases in psychological studies, experiments
are subjective; it is up to the experimenter to decide what the phenomena are. Ex: when you study
children and their interaction, it’s hard to decide just what is a friendly interaction.
Solution: make a precise description of what you are looking for; define the criteria. Also, build an
interrater reliability. This can be done in a subjective study by having two or more observants making
independent records. They compare the data, and when they are similar and agree on the same results,
then there is a high reliability that the results are right.
Random assignment is a way to create a group with the least number of confounding variables, since
there is little chance that the participants will have similar situations and backgrounds to them.
However, confounding vari