Chapter 2 The Ways and Means of Psychology
Case: Justine attempts to increase productivity at her family business. She notices that the chairs cause ache and
all so she switches it for an adjustable chair. She encourages the workers and the productivity increases by 20%
in one week. After a few weeks passes she stops by again and see that productivity has decreased by 20%. She
reintroduces the old chairs and works with the employees to her surprise productivity increase by 20%.
Scientific method allows us to determine the causes of phenomena. In psychology the goal is to
discover the causes of behaviour.
There are three basic forms; naturalistic observation, experiments and correlational studies. They
occur in a progressive sequence each one providing compelling evidence.
Natural observations – the observation of people and animals in their natural environment (little to
Correlational studies – can be observed in nature however they undergo formal measurements
Experiment – a study in which the researcher is able to manipulate the variable (independent) to see
the effects on the outcome (dependent variable). Experiments can confirm cause and effect relations
The five rules of experimentation
1. Identify problem and formulate hypothetical cause and effect relations among variables
2. Design the experiment
3. Perform the experiment
4. Evaluate the hypothesis by examining data obtained
5. Communicate results – scientist report in detail so that the experiment can be replicated
and replication ensures that there were no erroneous results in previous experiments
Hypothesis (Greek word for suggestion) statement about a cause-and-effect relation between 2 or
Theories are a set of statements that explains a known fact. Good theories generate testable
Variables can assume several values. An experimental group has expectation (the variables can be
manipulated by researcher. On the other hand a control group has no expectations (initial value at
zero). An independent variable can be manipulated and the dependent variable is measured.
Nominal fallacy should be avoided; explained by simply naming it (event). Naming it causes us to
cover events that cause behaviour. For example if a man shouts, saying that he’s angry doesn’t
explain why he’s acting that way.
Operational definition – definition of a variable in terms of the operations the researcher performs to
measure. Experiments require specific variables, we define our concepts operationally, in terms of
variables that can be measured or manipulated.
Validity – the validity of an experiment refers to how well the operational definitions (some variable)
capture the concept of interest.
Converging evidence strengthens the validity of an experiment. It is when someone has done the
same study before using different operational definition of the same concept and concludes the same
Confounding variables – confound means fail to distinguish. Inadvertent simultaneous manipulation
of more than one variable. The results of an experiment involving confounded variables permit no
valid conclusion about cause and effect. For example: there was a debate that argued whether children learn to read better by breaking down words (sup/er/man) or learning them as a whole
(superman). They found that the phonics way of learning to read worked much better. However right
before the research was conducted children would watch the show “sesame street” which used the
phonic way of teaching. This factor influenced the results therefore the results are inconclusive.
Counterbalancing – helps avoid confounds. A systematic variation of conditions in an experiment
such as the order of presentation of stimuli, so that different participants encounter them in different
orders; prevent confounding of independent variables with time-dependent processes such as
fatigue. (When testing which read method works better there should be two school which learn them
in different orders one learns breaking it down first while the other learns the whole word then you
switch this minimizes confounding variables to ensure that the students are not reading better
simply because they’ve already learned to read in another way before. To explain better if one had
already learned to read one way it should already be easier since they’ve grasped the basic concept of
Reliability – the repeatability of a measurement; the likelihood that if the measurement ware made
again it would