Questionnaires are an
easy method to gain
data,as the P's record
their own answers.
They also use a range
of open and closed Q's
to gather the data.
Open Q's: Open Q's allow
the P's to record their own
answers and to expand upon
why they think something &
give their opinion.It can also
be used as a way of
expanding upon an answer
given in a closed Q. Open
Q's tend to produce
qualitative data, which
means when analysing, you
look for trends in the
answers.
Closed Q's provide limited
answer choices and provide
quantitative data and the
answers are easy to analyse.
Interviews
Two types of
interviews:
Structured and
unstructured:
Structured interviews
use pre-determined Q's
that are restricted in
terms of answers.
Unstructured interviews have
no pre-determined Q's.
Likert scales
Likert scales can be
used to assess the
strength of a
person's opinion.
They are also an
example of a closed
question and provide
quantitative data.
However, when using a Likert scale,
P's may avoid the extreme ends of the
scale and only use the middle value,
thus not displaying their true opinion.
Experiments
The IV and DV:
The IV is the variable you are
deliberately changing/controlling and the
DV is the outcome of this change. From
this we can establish a cause and effect
relationship.
Controls
The experimenter controls the
IV to see if changes in the IV
lead to changes in the DV.
Hypotheses:
Experiments
use a
hypothesis to
"predict" the
outcome of the
experiment. A
hypothesis is a
testible
statement and
states that
there will be a
difference
found.
The opposite of a
hypothesis is the null
hypothesis, which a
'statement of
nothingness': i.e
"There will be no
difference"
Experiments use an
experimental
hypothesis, while every
other study uses an
'alternate hypothesis' as
they are not experiments
Three types of experiments: A lab
experiment manipulates the IV and uses
controls and takes place in a lab
environment. A field experiment still
manipulates the IV, but takes place in a
natural setting. A natural experiment has
a natually occuring IV that isn't
manipulated by the experimenter.
Strengths: From the data gathered we can assume a cause and effect
relationship and by manipulating the IV we can observe the effect it has on
the DV.
Weaknesses: P's are in an artifical situation which may lack realism
and P's may respond to the experimenter cues. This may mean the
results can't be generalised.
Correlations
A correlation is a way of measuring the relationship between two
variables. A positive correlation is when the two variables increase
together. A negative correlation is when one variable increases
while the other decreases. A zero correlation is when there is no
relationship between the two variables.
Strengths of correlations:
Correlations can be used when it
would be unethical to conduct an
experiment. If the correlation is
significant, then this justifies
further investigation. If the
correlation isn't significant, then
you can rule out a casual
relationship.
Weaknesses of correlations: Data
can be mis-interpreted and
assume a cause and effect
relationship, where it isn't possible
to do so; As you can only find a
link between the two variables.
There may be other unknown
variables that explain why the
co-variables being studied are
linked. Extraneous variables can
lead to a false conclusion.
A scattergraph is a graph that shows the
relationship between two sets of data
(co-variables) by plotting dots to represent
each set of scores. For each individual, we
obtain a score for each co-variable.
The extent of a correlation is described used a
correlation co-efficient. This is a number between +1
and -1. +1 is a perfect positive correlation and -1 is a
perfect negative correlation.
By using a correlation, we can assume
a link between the two variables, but we
can't assume a cause and effect
relationship.
Designs: For experiments, designs are used to
organise the P's in the study. Typically, there are
three main ways:
Independent measures: This is
when two different groups are used
in two different conditions.
Repeated measures: This is when the
same group of P's are used in two or
more different conditions.
Matched pairs: As independent measures but
the P's have been matched according to the
variable that is being investigated.
Observations
Time sampling: The
observer keeps a
count of each time a
particular event is
displayed.
Event sampling: The
observer decides on a time
interval, such as once a
minute. At the end of the
time interval, the observer
notes any particular
behaviours that are being
displayed by the target
individuals.
Unstructured observation: The researcher records all relevant
behavior but has no system. The behaviour to be studied is largely
unpredictable. One problem with unstructured observations is that
the behaviours recorded will oten be those which are most visible
to the observer and may not be the most important.
Controlled and Naturalistic: A controlled observation has some
variables controlled by the researcher such as the environment
and the specific behaviour being displayed,e.g a task. A
naturalistic observation has no variables controlled and is
studied in a natural environment.
Structured observations: The observer breaks up the
behaviours into categories and then uses the checklist to
know what catergories of behaviour to observe.
Participant and non-participant: In a participant observation, the
observer is merely watching the P's and acts as a non-participant. In
a participant observation, the observer becomes part of the group
that is being observed and acts as a participant and observes from
within the group
Strengths of observations: We can observe naturalistic
behaviour i.e "normal" behaviour. This is because the
behaviour has eco-logical validity. We can also gather data
firsthand and the observations can form the basis of future
investigations. It can also be used to check what people say
against what they do. It is also quick, easy and cheap to do.
Weaknesses of observations: There
may be observer bias, i.e the observer
may be subjective with what they see-
"only seeing what they want to see" i.e
based on their own interpretation of the
behaviour and this can affect the
validity. If the categories are unclear,
then we may get mis-recorded data
and also, ethical concerns- is it okay to
observe someone without their
consent?
Reliability and Validity
Validity: This is to the extent to which the research
has measured what it intends to measure.
This is when two different
measures are used to measure
the same thing. If the same scores
and results are obtained, then
concurrent validity is present.
Eco-logical validiy: Does the research
reflect real life behaviour, and real life
condititons?
Reliability: Reliability is whether the study and the
results are consistent, i.e If the same or very similiar
pattern of results are gathered, then the study and the
results are reliable.This is called test-retest.
Reliability can be improved by
having two researchers gather
the same data and results when
doing the same study. To gain
high inter-rater reliability, then
80% of the total data has to be
agreed that it is the same.
Case studies
A case study is a research
investigation that involves the
detailed study of an individual,
institution or event. It uses
information from a range of
sources, such as interviews and
observations, to obtain data.
These findings are often selected
and organised, for instance to
represent the individual's
thoughts and emotions.
Strength of case
studies: You can gather
rich and in-depth
information and it allows
you to see a change in
behavior over time.
Weaknesses of case studies: There
might be attrition in the sample and the
experimenter may lose objectivity in the
process. Also the results gained can't be
generalised to anyone else, as the
findings are unique to the person(s)
being studied.
Sampling: When studying any behaviour, researchers can't test everyone, so
they select a sample from the target population; i.e the people they are
interested in.
Opportunity sampling: This is the easiest
method, as you basically use people who
are readily available, i.e the first people
you find; however it is a biased option,
because the sample is drawn from a small
part of the target sample.
Self selected sample: This is
where the P's volunteer them
selves, usually be responding to an
advert. By using this sample, there
is access to a variety of P's, but the
sample is biased because the P's
are more likely to be motivated to
participate.
Random sampling: This is where the P's are selected randomly-i.e
with no bias. E.g Names of the target population into a hat and
draw out the required number. This sample is unbiased and all P's
have an equal chance of participating. However, may end up with a
biased sample, as not all P's identified will participate. It is also
almost impossible to carry out a random sample unless the sample
is very small.
Systematic sampling:
This is when you select
every N'th person from a
list.