Chapter 2: Psychology as a Science (09/13/13)
What is a Science?
-The chaos theory states that anything can happen at anytime…
-There are 2 core beliefs in science:
1. The universe operates according to natural laws
2. Such laws are discoverable & testable
-The scientific method uses logical reasoning derived from philosophy:
-Sir Francis Bacon=emphasized avoiding biases; argued that people look
selectively to confirm their beliefs & are thus subjective.
-Karl Popper: stressed that a good & sound theory must be falsifiable/be able to
be proven false. Theories has to be set up as either/or proposition, there has to
be a binary aspect, ie. Either right or wrong. Eg. Ryerson students are better or
nicer than UofT students (falsifiable)
*Freud says that if you ask a patient if they are thinking about sex and they say yes, then
it is true but if they say no, that means they also thought about it just by listening to the
question. There was no falsifiability in his theory b/c he was right in both cases.
-Steps in the scientific method (see fig 2-1 pg.32)
-Operational definitions: an operational defn is how we (the researcher) decide to
measure our variables in order to express hypothesis in a clear, specific & concrete
manner. There are usually many ways to measure a variable, eg. aggression,
depression, sense of humour. When you do research, you have to decide how you’re
going to measure the variables.
-Choose participants: 1) Population—the entire group of interest. 2) Sample—a portion
of the popn that is selected for the study. 3) Random selection—randomly choosing a
sample from a popn; in an ideal world, random selection is the best way to choose a
sample but is rarely used in psychology. 4) Sampling bias—choosing a sample that
doesn’t represent your popn, eg. offering compensation for people’s time to volunteer in
the study is a recruitment strategy to get a volunteer sample is subject to sampling bias
b/c ppl w/ financial need might be more likely to volunteer, people w/ unpleasant
experience in class, people who are interested. These factors are subject to sampling
bias which will yield to different results than you would normally get if a random selection
was done. *No matter what method of choosing participants, you can never completely
get rid of sampling bias.
-There are 2 basic types of research:
1. Descriptive research=used to observe & describe behaviour & to determine
the existence of a relationship b/w variables. Nothing is manipulated and it
doesn’t enable you to deduce causality, eg. the more classes one has to take the
less rewarding there classes are; this doesn’t show a cause instead it merely
describes the relationship.
a. Case studies-an intensive study of 1 or 2 people, eg. conjoined twins
who thrived into adulthood. Advantages include: the only method you can use if
the type of behaviour you’re looking at is rare. Disadvantages incl: cannot
generalize results to fit the whole popn.
b. Naturalistic observation-systematic method of people watching. You get
more of a sense of reality w/ this method than merely reading reports, eg. people
on the ttc would report that they would offer their seat to someone who needs it
but if you observe the same ppl personally, there is a huge discrepancy in the
numbers b/w those who said yes and those who actually did it. Advantages incl:
can study things that are too unethical for an experiment or that people might lie
about. Disadvantages incl: can’t determine causality & presence of the
Hawthorne effect (if people know they are being watch they tend to change their behaviour for the better).
c. Surveys-asks people directly about their behaviours; in topics like drug
use/sexual activity, you can still use a survey but you have to be careful that the
participants are not threatened to have their identity revealed. Surveys just
present associations of things at one pt in time that are not necessarily causally
related, such that there may be other/confounding factors, eg. extreme heat
aggravate aggression which results in correlation seen b/w ice cream & crime
rates. Advantages incl: data collection is quick. Disadvantages incl: sometimes
people don’t tell the truth & can’t determine cause & effect.
*All descriptive research methods don’t determine causality.
2. Experimental=used to demonstrate causal relationship b/w variables.
Experiments examine how one variable causes another variable to change. You can
bring people in laboratory settings, have them do something to get causality but the
results may not reflect real life thus you can’t generalize. Advantages include: can
establish causality. Disadvantages incl: might not be generalizable to real world
situations (outside of lab). There are 2 kinds of important variables in a study: 1)
independent variable (IV)—one you can manipulated & what the researcher changes to
see what happens. A punitive cause, eg. is egg important in cookies? You can do a study
involving cookies w/ eggs & cookies w/o eggs. What is the cause? 2) Dependent