Psychology – Chapter 11

Research Methods Notes

Date: 09/11/2024

Variables-
Independent Variable - the variable that is directly manipulated by the researcher

Dependent Variable - the variable that is being measured in the study

Extraneous Variable - a variable that is not controlled but could affect the results of the
study

Confounding Variable - an extraneous variable that affects the results of the study so
the effect of IV is not clearly seen.

Situational Variable - an extraneous variable present in the environment of the study

Participant Variable - extraneous variable specific to the participants of the study ( ex -

mood )

—------------------------------------------------------------------------------------------------------------------------------------------

Order Effects - when the participants improve or worsen in the second condition due to
fatigue or practice.

Demand Characteristics - when the participants alter their behaviour and act desirably
according to the perceived aims of the study.
 1

Investigator effect - when a researcher unintentionally gives clues to the participant,

altering their behaviour.

Controlling Extraneous Variables -

Situational Variables -

- Standardised procedure - when the procedure of the study is the same across all
conditions.
- Counterbalancing - where half of the group experiences condition A then B and
the other half experiences condition B then A ( to remove order effects ).
- Randomisation - when participants are randomly assigned for condition A or B as
their first or second ( to remove order effects ).
- Single-blind Technique - when the study’s aims/information is withheld from the
participants
- Double-blind Technique - when the study’s aims are withheld from the
participants and the researcher.

Participant Variables -

- use the same participants in both conditions of the study, so the data for each
participant can be compared across both conditions.
- different participants but make sure they are matched on important
characteristics.
- Random Allocation - when participants are randomly assigned to a condition of
the study.

Hypothesis
1- Null Hypothesis
Predictions that will fail to show any difference or relationship that is consistent.
Ex - There is no difference between the number of digits recalled and whether
participants listened to classical music or jazz. Any difference is due to chance.

2- Alternate (experimental ) Hypothesis

- prediction of the outcome of a study based on what is expected to happen.
Directional hypothesis: a hypothesis that predicts the direction the results will go in.
 2

Non-directional hypothesis: a hypothesis that predicts that a difference/relationship will

be found, but does not specify what the difference/relationship will be.
Experimental Hypothesis: the name given to a hypothesis when used in field and
laboratory experiments.

Sampling
1. Random sampling technique - all have an equal chance of being selected (pick
from hat)
2. Stratified sampling technique - a technique that ensures subgroups of the target
population are proportionately represented in a sample.
3. Volunteer sampling technique - a technique that asks for participants by placing
an advert for volunteers.
4. Opportunity sampling - a technique that recruits participants who are readily
available at the time.

Strengths and Weakness -

Sampling Strengths Weakness

Random An equal chance of being selected - Some may refuse to participate

helps control participant variables as this is ethically in their rights,
- Unbiased which can lead to a less
representative sample.
- time-consuming

Stratified Completely representative of the - Time-consuming as

target population proportion calculations
- People can refuse ->
sample error

Volunteer Minimal effort Less representative

Most ethical sampling- participants Biased sample
directly asked.

Opportunity Quickest and simplest Biased sample because not

every member of the target
Saves time and effort population may be around at the
time of the study.
 3

Experimental Designs -
1. Independent measures design - participants are split into groups, with each
group tested in only one condition of a study. ( CONTROL - RANDOM
ALLOCATION )
2. Repeated measures design – this involves using the same participants in all
conditions of a study. ( CONTROL ORDER EFFECTS - COUNTERBALANCING OR
RANDOMISATION )
3. Matched pairs design: different participants are used in each condition of the
study, but are matched for likeness on important characteristics.

Strengths and Weakness -

Designs Strengths Weakness

Independent Measures There are no order effects as More participants are

participants only take part in needed than for a repeated
one condition of the study. measures design.
Individual differences

Repeated Measures Fewer participants are Demand characteristics

needed, making it more are more likely as
economical. participants are more able
No individual differences to guess the aim.
Order effects- results may
reflect practice or fatigue.

Matched Pairs Fair comparisons can be It is time-consuming to

made between the groups as match participants and not
they are equally matched. all characteristics can be
equally matched.

Reliability and Validity

Reliability: the consistency of an outcome or result of an investigation (a measure).
Validity - if the test measures what was intended to.
Internal validity: whether the measures used in a test genuinely test what they were
designed to test.
External validity: whether the findings are generalisable to the target population
 4

Qualitative methods - gathering lots of detailed information. Qualitative methods can

include case studies, unstructured interviews and participant observation. ( hard to
replicate so may not produce reliable findings and lacks generalisability. )
- researcher bias(when a researcher interprets the outcome of a study according to
their own view (subjective) and investigator effects
- Triangulation: when more than one measure is taken for a behaviour to
cross-validate the findings.

Quantitative methods - close-ended questionnaire/surveys

gathering numeric data that can be used for graphs and statistical data.

Ethical Issues
BPS ( British Psychological Society ) CODE -
1- Respect for the autonomy, privacy and dignity of individuals and communities
2- Scientific integrity
3-Social responsibility
4- Maximising benefit and minimising harm

- Informed consent
- Deception - misleading or lying to participants about the nature of the
investigation. -> debrief
- Confidentiality- participant information should be kept confidential at all times to
ensure participants cannot be identified by the research.
- Right to withdraw
- Protection of participants - physical and psychological harm

Experiments -
1- Laboratory Experiment - an investigation that takes place in a controlled environment.

- controlling or eliminating other extraneous variables (researchers manipulate IV)

2- Field Experiment - an experiment conducted in a natural environment

 5

- (researchers manipulate IV)

- control over extraneous variables, but the level of control is not as high as in a
laboratory because naturalistic environments are more unpredictable.

3- Natural Experiment - occurs in a real-life environment and has an IV and DV, but the
researchers themselves do not manipulate the IV as it occurs naturally

Strengths and Weaknesses -

TYPE STRENGTHS WEAKNESSES

Laboratory Scientific - high levels of control Unnatural environment- lacks ecological validity
Internal Validity - cause and effect Participants invited so may show demand
detected characteristics
Reductionist as it doesn't involve other variables

Field Natural environment so natural Not all extraneous variables are controlled as it’s
behaviour a natural environment
Internal Validity - cause and effect are Participants may have been unaware so
established and some extraneous researchers didn’t give informed consent.
variables are controlled

Natural The IV is naturally occurring and not It is not possible to randomly allocate
manipulated. participants to conditions of the IV- issues with
The environment in which participants participant variables
are studied is real and naturalistic. Extraneous variables are difficult to control

Self Report Data

Interviews

1. Structured: the questions are already pre-set to ask interviewees.

2. Unstructured: a free-flowing conversation around a particular topic with a
respondent.
3. Semi-structured: a mix of pre-set questions and unprepared questions asked to a
respondent.

Questionnaires -

1- Closed-ended questions: questions with a fixed response to choose from.

2- Open-ended questions: questions with no fixed response.

 6

Interviewer effect: the characteristics of an interviewer impact the way a respondent answers
questions.

Strengths and Weaknesses

Strengths Weaknesses
INTERVIEW Structured- Social desirability bias, where
gather a lot of information from a large respondents answer in a way they feel
number of people relatively quickly. is more socially acceptable.
very little training Interviewer effect.
Easy to replicate Structured- can’t answer freely
Semi-structures Semi-structures - more training and
Freedom to answer time-consuming than structured
Unstructured Unstructured- well-trained,
Most depth information time-consuming and hard to replicate

QUESTIONNAIRE Inexpensive - large sample The response rate for questionnaires is

Ethical - people choose to answer relatively low, as people may not have
No interviewer effect the time or inclination to complete and
Reduce the social desirability effect. return the questionnaire.
Closed-ended questionnaires produce Close-ended - right option is not
quantitative data, which is relatively easy available to them or they would like to
to analyse and is objective. explain their choices.
Open-ended - freedom for respondents Open-ended - difficult to analyse

—------------------------------------------------------------------------------------------------------------------

Correlations -
- Look for relationships between variables
Positive correlation: as one co-variable increases, the other co-variable increases.
Negative correlation: as one co-variable increases, the other co-variable decreases.
Strengths - cheaper and less time-consuming / investigate a topic that's difficult to study / less
interventionist and more ethical/objective as they show quantitative data
Weakness - cause and effect cannot be established / lack detail
Case Study -

- in-depth investigation into one person or a group of people.

- generally suitable for the study of unique or rare individuals
 7

- could not have been staged for ethical or practical reasons.

Strengths - unethical to set up experimentally / it provides the most in-depth information
Weaknesses - cannot be repeated to check the reliability of the findings / may not be applicable
to people and so they lack generalisability / interpret the findings differently from others. This is
known as researcher bias( but it can be avoided if triangulation is used – a researcher uses
different methods to validate their findings )

Observations -
a research method that involves watching and recording behaviour.

- Naturalistic observation: an observation conducted in an everyday environment

where the behaviour being studied is normally seen.
- Controlled or structured observation: an observation carried out in a laboratory or
controlled environment.
- Overt observation: participants know they are being observed as part of an
investigation
- Covert observation: participants are unaware that they are being observed.
- Participant observation: when an observer is involved in the group they are
observing.
- Non-participant observation: the observer watches and records people without
being actively involved.

( Inter-rater reliability: when more than one observer codes behaviour and their results
are compared to check for agreement )

Strengths - overt observations are ethical / most reliable observation is structured

Weaknesses - lack validity - observer bias ( when an observer interprets the observed
behaviour according to their own view )

- There may be ethical issues associated with observational research. Covert

observations can raise the issue of lack of consent. Participant covert
observations raise particular ethical concerns because the researcher, who may
be acting as a group member, deceives participants.

Representing and Interpreting data -

Normal distribution: when mean, median and mode are very similar or the same.
 8

Skewed distribution: when median and/or mode differ from the mean.

Frequency diagram/histogram: illustrates frequency to show the distribution of

continuous data.

Bell curve: the shape of a normal distribution curve.

Types of Data -
- Primary data: data collected directly for a specific research purpose.
- Secondary data: data used in a study that has already been collected,
often for a different purpose but similar.
- Qualitative data: data that are descriptive, not numbers, such as words or
pictures ( in-depth and detailed )
- Quantitative data: numerical data that can represented in graphs/
statistics
 9

( Meta-analysis: a procedure used to merge and analyse findings from studies focusing
on a similar issue in order to draw overall conclusions )
 10

Issues and Debates -

Ethical issues

5 ethical issues: informed consent, deception, confidentiality, the right to withdraw

and the protection of participants. Morality is also an ethical issue.

1- Development - ( Minimise harm and protect participants )

Children should be researched ‘with’ rather than researched ‘on’; Piaget, Dweck and
Willingham tended to focus on doing research ‘on’ children

Young people tend to be more participatory, which means it involves the children in the
research process.

( Participation rights: the rights of people, including children, to participate in decisions

about them. Protection rights: the rights of a child to be protected at all times. )

2- Memory - ( deceit / informed consent / no debrief )

Peterson and Peterson (1959), for example, had told participants that the longer the
interference task the lower the recall, participants might have acted according to what
was said and results might have been affected.

If Bartlett (1932) had explained that he expected poorer recall of the story over time and
alterations of the story so it made more sense, this also could have affected the results.

Research proposal submitted and should be approved by an appropriate ethics

committee.

3- Psychological Problems - ( nature of the topic and the participants being studied )

Young - waiting list or a control group - these participants don't get therapy - ethics

Caspi - people cannot help or change their genetic make-up / very stressful and
upsetting
 11

4- Brain and Neuropsychology ( protection of participants an informed consent from

brain damage participants )

Case studies where the individual is vulnerable to the brain damage that is being
researched.

Confidentiality, another ethical issue, was given. HM’s name was not used until after his
death.

Treatment should always be the main focus.

Researchers who look at the brain and neuropsychology have a responsibility to

participants and to society regarding their findings, and this responsibility is an ethical
issue.

5- Social Influence - ( protecting participants

under’ social influence itself carries ethical questions.

Milgram - brutality and from knowing how far they themselves might go.

Zimbardo - guard brutality / not show someone a side of themselves that is scary for
them and they should not harm participants.

The ends justify the means

Risk-benefit: comparing a study’s risks and benefits to ensure the benefits outweigh
the risks.

- If there is the right to withdraw at any time, a full debrief, a conclusion that the
benefits will outweigh any harm and that any harm is temporary, then deception
can be justified in ethical terms.