🔎
Research methods
🔑Key terms
extraneous variables - variables that may have an effect on the DV if not
controlled
confounding variables - a variable that wasn’t the IV that affected the DV
demand characteristics - cues from the researchers that lead to participants
guessing the experiment objective and changing their behaviour
investigator effects - where the investigator’s behaviour has an effect on the
research outcome
quantitative data - numerical data
qualitative data - information that can be recorded but is not numerical
🔍Aims and hypothesis
aim: a statement that says what the researcher intends to investigate
hypothesis: an idea that can be verified or disproven by an investigation
Research methods 1
directional hypothesis (one-tailed) - gives a direction to the prediction
e.g. ‘women perform better than men in exams’
non- directional hypothesis - does not give a direction to the prediction
e.g. ‘there will be a difference in exam performance between men and
women’
null hypothesis - states that there is no relationship between the variables being
studied
falsifiability - suggests that for a theory to be considered scientific, it must be
able to be tested and proven as true or false
🫂Sampling
sample - the participants you select from a target population to make
generalisations about
representative - the extend to which a sample mirrors a researcher’s target
population
generalisability - the extend to which the findings can be applied to the larger
population
volunteer sample - where participants pick themselves through newspaper
adverts or through the internet
convenient
unrepresentative - leads to bias on the part of the participant because those
who volunteer have particular characteristics (like confidence)
random sampling - where every member of a population has an equal chance of
being selected
for large samples it provides the best chance of an unbiased sample
time consuming to create a list of every individual to choose from
Research methods 2
stratified sampling - dividing the target population into subcategories and
selecting members in the proportion that they occur in the population
aims to make the sample representative
can be time consuming to create sub-categories and calculate proportions
systematic sampling - chooses subjects in a logical way (like choosing every nth
participant on a list of names)
aims to make the sample representative
difficult to achieve this
opportunity sampling - using people from the target population that are available
at the time and willing to participate
quick and easy way of choosing participants
may not be representative
❓Variables
independent variable - the variable the experimenter manipulates that has an
effect on the dependent variable
dependent variable - the variable the experimenter measures after making
changes to the IV
operationalisation of variables - ensuring variables are in a form that can be
tested
extraneous variables - anything that is not the IV that has the potential to affect
the results
confounding variables - an unmeasured third variable that influences the
relationship between the IV and DV
happens when an extraneous variable has not been properly controlled in
the experiment
Research methods 3
🖌Experimental design
independent measures - where different participants are used in each condition
of the independent variable
more participants are required
avoids order effects (like practice and tiredness)
prevents demand characteristics
repeated measures - where the same participants take part in each condition of
the IV
participant variables (like individual differences) are reduced because the
same participants are used for all conditions
less participants required
may be order effects
can use counterbalancing
matched pairs - where pairs of participants are matched in terms of key variables
and one member of each pair does each condition
reduces participant variables
avoids order effects
time consuming to find closely matched pairs
if one person drops out, you lose 2 peoples data
impossible to perfectly match people
🥼Types of experiment
lab experiment - conducted in a wall controlled environment (may not be a
laboratory)
Research methods 4
allows objective measurements to be used
easy to replicate
standardised
allow for control of variables
high internal validity
low ecological validity
field experiment - where the experimenter manipulates the independent variable
but in a real life setting
more likely to reflect real life
higher ecological validity than lab experiment
less demand characteristics (if covert)
less control over extraneous variables
natural experiment - where the experimenter conducts the experiment in the
everyday environment of the participants and does not control the IV
more likely to reflect real life
high ecological validity
can be used in situations where it is unethical to manipulate the IV
less likely to have demand characteristics
no control over extraneous variables
more time consuming to conduct
quasi experiment- where there is a naturally occurring independent variable that
is investigated
e.g. gender
the IV is a difference between people that already exists
high internal validity
Research methods 5
hard to account for all confounding variables
case studies - in depth investigations of a single person or specific group
provides qualitative data (words)
provides detailed information
allows investigation of otherwise unethical situations
lack scientific rigour
researcher bias
hard to replicate
time consuming
🗣Interviews (self report)
structured interview - where the interviewer asks a set of prepared, closed
questions that are read out exactly as worded
standardised
easy to replicate
not flexible so we may miss out on insightful information that someone
could’ve provided
generates quantitative data
unstructured interview - where the interviewer asks open ended questions and
lets the interview flow like a natural conversation
questions can be modified to suit each participant
produce qualitative data
allows the interviewer to ask for clarification or more detail
time consuming
Research methods 6
semi-structured interview - contains pre-determined questions but also allows for
informal chat
maintains the validity from structured interviews whilst also allowing the
interviewer to ask for elaboration
interviewer has to be well trained
📄Questionnaires (self report)
where participants are asked to answer a series of questions
relatively cheap and efficient when obtaining lots of information from lots of
people
people may lie due to social desirability bias (wanting to look good)
acquiescence bias (agreeing to everything to get it over with quicker)
avoids investigator effects
closed questions - where there is a fixed number of responses
like tick boxes, Likert scales, rating scales
collects quantitative data which is easier to analyse
lack depth and detail
open questions - where respondents are free to answer how they wish
collects qualitative data which is harder to analyse and compare because
the answers vary
more in depth and in detail
🔎Types of observation
Research methods 7
naturalistic - where the researcher watches the behaviour in an environment that
is naturally occurs in
less controlled environment
higher ecological validity
controlled - where the researcher watches and records behaviour in a structured
environment
allows variables to be manipulated
increases internal validity
lower ecological validity
covert - where the researcher records behaviour without the participant’s
knowledge or consent
removes demand characteristics
ethical issues
overt - where the researcher records behaviour with the participant’s knowledge
and consent
ethical
increase in demand characteristics
participant - where the researcher is a member or the group being watched
increases insight into participant behaviour
loss of objectivity if researcher becomes too involved
researcher effects
non-participant - where the researcher remains outside of the group being
studied
remains objective
Research methods 8
lose valuable insights
👀Observations
behavioural categories - where the target behaviour is broken down into smaller
parts that can be checked and measured
provides a structured record
must ensure that all researchers interpret behaviours in the same way
takes a long time to properly develop behavioural categories
inter-rater reliability - where observations are carried out by two or more
observers, the observers need to make sure that they are being consistent in
their judgements
can reduce bias
effective with behavioural categories
continuous sampling - recording all instances of the target behaviour
event sampling - where the target behaviour is pre-established and every time
the behaviour is observed, it is recorded
useful for frequent behaviours that would otherwise be missed
complex behaviours may be reduced to one aspect
time sampling - recording what is seen within a fixed time frame
reduces the number of observations that are made
may be unrepresentative of the whole observation
Research methods 9
🧪Research design
cross-sectional research - where the researcher compares multiple segments of
the population at the same time
longitudinal research - where data gathering is done repeatedly over an
extended period of time
cohort studies - where the participants share a common factor or characteristic
⭐Reliability
reliability - a measure of consistency
test-retest = assessing the same person on two different occasions to see to
which extent the test produces the same results
inter-rater reliability = the extent to which there is an agreement between two or
more observers
📂Meta analysis
a systematic review that involves identifying an aim and searching for existing
studies that have addressed similar aims
increases validity of the conclusions
research designs vary so are not truly comparable
👭Peer review
where research is checked over and analysed by other researchers
feedback from the reviewer determines whether or not the article is accepted
and published
Research methods 10
prevents faulty data from entering the public domain
slows down publication of materials
some reviewers use it as an opportunity to prevent competing researchers from
publishing their work
✅Types of validity
concurrent validity - the extent to which a psychological measure relates to
existing similar measure and obtains close results
face validity - does the test measure what its supposed to measure
ecological validity - the extent to which findings can be generalised to other
settings
temporal validity - the extent to which findings from a study can be generalised
to other time periods
💯Statistical testing
significant result - where there is a low probability that chance factors were
responsible for the results in the variables tested
if the test is significant - reject the null hypothesis and accept alternative
hypothesis
if the test is not significant - reject alternative hypothesis and accept null
hypothesis
type 1 error - when the null hypothesis is rejected when it should’ve been
accepted
type 2 error - when the null hypothesis is accepted when it should have been
rejected
Steps to the sign test
Research methods 11
aim: to find out whether student prefer a cooked, non-cooked breakfast or
neither
15 participants
non cooked breakfast = -
cooked breakfast = +
1. Add up the number of positives and negatives
2. Add up the number of times the less frequent sign occurs (S)
3. N = the number of participants (minus any where the results were omitted)
4. determine whether it is a one tailed or two tailed hypothesis
5. use the critical value table
6. observed value must be lower than the critical value for the result to be
significant (unless stated otherwise)
Choosing a stats test
Research methods 12
ordinal data - where the data is ordered but the intervals are not equal
nominal data - categorical data where the item only appears in one category
interval data - uses standardised scales of measurement
⚖Ethical issues
informed consent - when the participants are able to make an informed
judgement about whether to take part in the experiment
deception - involves deliberately misleading or withholding information
right to withdraw - participants can leave the study and also request for their data
to be removed from the research
protection from harm - risks greater than those experienced in everyday life
should be avoided
confidentiality - concerns the communication of personal information
Research methods 13