Research Methods

Experimental method

manipulation of IV to have an effect on the DC which is measured and stated in results (field, lab, quasi, natural)

Aim

General statement about what the researchers tend to investigate (purpose of the study)

Hypothesis

A clear, precise testable statement that states the relationship between variables in the study. (2 Conditions)

Non-directional hypothesis

Does not state the direction of the difference or relationship

1. 2 Tailed

2. There will be a difference in DV between condition 1 compared to condition 2

e.g. the difference in the amount of hours of sleep a participant has will have an effect on their memory performance which is shown in their test scores.

Directional hypothesis

States the direction of the difference or relationship (usually includes words such as more, less,, higher, lower)

1. 1 Tailed

2. Participants who condition 1 will get higher/lower DV compared to condition 2

E.g. the more sleep a participant has, the better their memory performance

When to use directional hypothesis

When a theory/findings of previous research suggest a particular outcome

When to use a non-directional hypothesis

When there is no theory of previous research or findings that contradict it

Null hypothesis

no difference

1. There will be no difference in DV between condition 1 and condition 2

Variable

Any thing that can very or change within an investigation and determines if changes in one thing results in changes in another

Independent variable

Aspects of experiment that can be changed or manipulated to have an effect on DV

Dependent variable

Aspects of experiment which is measured

Control variable

Aspects of experiment which stay the same

Operational variable

Aspects of experiment that can be repeated

Operationalisation

Clearly defining variables in terms of how they can be measured

1. Variables defined + measurable

2. Hypothesis should show operationalisation

3. Number of Q’s

Experimental design

Different ways in which participants can be organised in relation to experimental conditions

1. Independent group design

2. Repeated measures

3. Matched pairs

Independent group design

Participants are allocated to different groups where each group represents one experimental condition. Example: one group drinks an energy drink and the other group water.

Independent group evaluation

1. Individual difference - no control over P’s variables as abilities in conditions can change DV (random allocation makes equal chance of being in a condition)

2. Less economical then repeated - each participant cornubites a single results only meaning more time and money is spent on recruiting

3. Order effects - not a problem as participants are less likely to guess aim of study

4. No Demand characteristics

Repeated measures

All participants take part in all conditions of the experiment

Repeated measures evaluation

1. Order effects - each participant has to do both tasks meaning first task may have a continuing effect when doing the second task (counterbalancing)

2. Fatigue - participants having to complete both tasks may lead to tedium and tiredness

3. Demand characteristics - participants may work out aim of study due to experiencing all conditions

4. Less time consuming - less P’s needed

Matched pairs

Pairs of participants are first matched on some variables that may affect the dependent variable. Then one member is assigned to condition A and the other condition B

Matched pairs evaluation

1. Less order effect + demand characteristics - only take part in a single condition

2. Participants can’t be matched exactly - there is still differences which will effect DV

3. Time consuming + Expensive - pre test may be required

Target Population

A group of people who are the focus of the researchers interests

Sampling

A group of people who take part in research. The sample is drawn from the target population and is presumed to be representative

1. Opportunity

2. Random

3. Systematic

4. Stratified

5. Volunteer

Bias (sampling context)

When certain group are over or under represented within the ample selected

Generalisation

The extent to which findings and conclusion form a particular investigation can be broadly applied to the population

Random sampling

Participant are chosen randomly through a “lottery method” so have an equal chance of being selected

eg picking from a hat

Random sample evaluation

1. Participants may refuse - leads to more of a volunteer sample + unrepresentative sample

2. Time Consuming - takes time

3. Potentially unbias - cofounding variables may be equally divided between the different groups which enhances internal validity

4. No researcher bias

Systematic sampling

Every nth member of target population is selected

Systematic evaluation

1. Objective - once system selection has been established, the researcher has no influence on who is chosen

2. Time consuming - takes time for system to establish causing tedium

3. Fairly representative

Stratified sample

Identifying sub-groups into categories and proportionally taking people from these sub-groups to make a representative sample

Stratified evaluation

1. Representative sample - accurately reflects the composition of population meaning generalisations become possible

2. Free from researcher bias - random selection

3. Time consuming

Opportunity sample

Participants who are willing or near to take part (recruited conveniently)

Opportunity sample evaluation

1. Time efficient - convenient

2. Unrepresentative - drawn from specific area so can’t generalise whole population

3. Researcher Bias - researcher has complete control over selection of participants

Volunteer sample

Participants are willing to self-select themselves

Volunteer sample evaluation

1. Time Efficient - participants go to researcher so no recruitment

2. Volunteer bias - may attract a certain profile of a person (affect generalisability)

3. P’s agenda - not take seriously (e.g. money)

Extraneous variable

Any variable other then the independent variable that may affect the dependent variable if not controlled (nuisance variables that make it harder to detect result)

e.g. lighting in the lab

1. Participant variables

2. Experimenter/ Researcher variables

3. Demand characteristics

4. Situational variables

Confounding variables

A kind of extraneous variable that varies systematically with the IV (difficult to be sure of origin of impact of DV)

E.g. time of day experiment done

Participant variables

Differences between participants that could influence results e.g. age, iq

1. Matched pairs

Experimenter/ Researcher variables

Factors that affect participants results e.g. researched gender, appearance

1. Double blind

Demand characteristics

Any cue from the researcher or research situation that may be interpreted by participants as revealing the purpose of investigation

1. P’s act differently

2. Participant reactivity

3. Unnatural behaviour impact valid

4. Single blind

Situational variables

Factors of environment that influence outcomes e.g. weather

Investigator effects

Any unwanted influence of the investigators behaviour on the research outcome (DV) e.g interaction with participants

Randomisation

The use of chance methods to control for order effects of bias when designing materials and deciding order of experimental conditions

Standardisation

Using exactly the same formalised procedures and instructions for all participants in a study

1. Eliminate EV

Laboratory experiments

An experiment that takes place in a controlled environment within which the researcher manipulates IV and records effect on DV, and maintains EV

1. Variables carefully controlled

Laboratory experiment Evaluation

1. High control over EV and CV - only IV has affect on DV creating high internal validity

2. Replication - high level of control allows repetition creating validity

3. Low Ecological validity - more artificial then everyday life creating low generalisability

4. Participants aware of testing - demand characteristics

Field experiments

Takes place in a natural setting within which the researcher manipulates IV and records effects on DV

Field experiments Evaluation

1. Higher Mundane Realism - produce more authentic behaviour

2. High Ecological Validity - naturalistic

3. EV - loss control so can’t replicate

4. Ethical Issues - participants are unaware of study so cannot consent leading to invasion of privacy

Natural Experiments

The change in IV is not brought by researcher but happens naturally and researcher on records effect on DV

Natural Experiments Evaluation

1. High External Validity - involve real world issues (ecological validity + generalisable)

2. Participants Not Randomly Allocated - researcher may be unsure whether the IV affected DV (EV + CV)

3. Naturally Occurring Event Happens Rarely - reduces opportunities for research and limits generalising findings to similar situations

Quasi Experiments

A study that is almost an experiment but lacks key ingredients as the IV has not been determined by anyone and the variables naturally exist eg being old or young

Quasi Experiments Evaluation

1. Controlled Conditions - can be replicated

2. Replicable

3. Participants Not Randomly Allocated - confounding variables

Informed Consent

Involves making participants aware of aims and procedures of research and what their data will be used for

(some researchers think this makes study meaningless as they won’t act naturally)

Dealing With Informed Consent

1. Consent letter - contains relevant information

2. Children require parental consent

3. Retrospective - consent after complete study

Deception

Deliberately misleading or withholding information from participants ,this means there’s no informed consent.

(some researchers believe this is justified as it avoids change in behaviour and undue distress)

Dealing With Deception

1. Debrief at the end of study - true nature

2. Right to withhold data

3. Aware of aims

Right To Withdraw

All participants should know they can withdraw from study at any time and also their data

Dealing With Right To Withdraw

1. Reminded through study

Protection From Harm

Participants should not be put at risk than they would be in their daily lives both physically and mentally eg embarrassment

Dealing With Protection From Harm

1. Counselling

2. Reassured about feelings

3. Cost benefit analysis

Privacy and Confidentiality

Participants have the right to control information about themselves and have their data protected.

(Researchers should avoid research in their private expectations and avoid identifying details)

Dealing With Privacy and Confidentiality

1. Anonymity

2. Reminded in debrief + brief

3. Participants are referred as numbers or initials

Pilot study

A small-scale version of an investigation that takes place before real investigation in conducted

Aims of a pilot study

1. Check procedure runs smoothly

2. Make any modifications

3. Try out questions and remove any too ambiguous

4. Stop potential issues

5. Save money and time

Single-blind Procedure

Researcher withholds aims of research from participants and conditions of experiment

1. Stop P’s bias

2. Avoid confounding effects of DC

Double-blind Procedure

Neither the participants nor the researcher who conducts the experiment is aware of aims of investigation.

1. Prevent bias

2. Prevent DC

3. Stop placebo

4. Reduce investigator effects

Control Groups

A neutral group which acts as a baseline to formulate comparisons with.

1. If the change in behaviour of the experimental group is significantly greater than that of the control group, the researcher can conclude cause of this was IV.

Observational techniques

1. Naturalistic

2. Controlled

3. Overt

4. Covert

5. Participant

6. Non-participant

Naturalistic Observation

Watching and recording behaviour in its natural setting in which it would normally occur

Naturalistic Observation Evaluation

1. High External Validity - generalised to everyday life

2. High Ecological validity - high mundane realism

3. Lack Replication - CVs and EV make it difficult

4. EV + CV - low internal validity

Controlled Observation

Watching and recording behaviour within a structured environment where variables are controlled and manipulated

1. Focus on particular aspect of behaviour

Controlled Observation Evaluation

1. Low Ecological Validity - low mundane realism

2. Demand characteristics - unnatural behaviour

3. CVs + EV controlled - easier observation (high internal val)

4. Replicable

Covert Observation

Participants behaviour is watched and recorded without their knowledge or consent

Covert Observation Evaluation

1. no demand characteristics - high validity + natural behaviour

2. Removes P’s Reactivity

3. (L) Ethics Questioned - no informed consent

Overt Observation

Participants behaviour is watched and recorded with their knowledge and consent

Overt Observation Evaluation

1. Ethically Suitable - consent given

2. Low Internal Validity - demand characteristics

Participants Observation

Researcher becomes part of group who is being observed

Participant Observation Evaluation

1. High External Validity - increased insight of lives of participants

2. Lose Objectivity - identify too strongly with participants ‘going native’

Non-participant Observation

Researcher remains outside of group who is being observed

Non-participant Observation Evaluation

1. High Objectivity - less chance of ‘going native’

2. Low External Validity - lose insight on participants

Observational designs

1. One issue is observer bias

2. Solve this with inter observer reliability

3. Unstructured

4. Structured

Structured observation

Researcher quantifies what they’re observing using predetermined list of behaviour and sampling methods

1. Simplify target behaviour using behavioural categories

2. Quantitative data

Structured Observation Evaluation

1. (S) Recording data is more systematic and easier

2. (S) Quantitative data is straight forward to analyse

3. (S) Less risk of observer bias

4. (L) Lack detail and richness of qualitive data

5. (L) Difficult to achieve inter observer reliability

Unstructured Observation

Continuous recording where researcher writes everything they see

1. Suitable for small scale experiments

2. Qualitative data

Unstructured Observation Evaluation

1. (S) Can provide rich information and detail in depth

2. (L) Observer bias (only record what you see)

3. (L) More difficult to analyse data

Behavioural Categories

When a target behaviour is broken up into components that are observable and measurable. This breaks target behaviour up into a set of behavioural categories.

What do behavioural categories need to be?

1. Operationalised - target behaviour should be precisely defined

2. Objective + Observable - no need for inferences

3. Cover all possible contents

4. Be mutually exclusive - don’t overlap when marking

5. Unambiguous

Structured interview sampling methods

1. Time

2. Event

Event sampling

Recording the number of an event/behaviour every time it occurs

Event sampling evaluation

1. Useful When Behaviour Is Infrequent - records behaviour which could be missed by time

2. Details May Be Overlooked - event could be too complex

3. Counting Errors - if skill is very frequent

Time sampling

Recording a behaviour/event in a fixed time frame that is pre-established before observation

Time sampling evaluation

1. Reduces number of observations - less need to be made and analysed + time consuming

2. unrepresentative - doesn’t reflect behaviour in whole observation

Self-report

Any method in which a person is asked to state or explain their own feelings, behavior, opinion and experiences on a given topic

1. Questionnaire

2. Interview

Questionaire

A set of written questions used to asses a persons thoughts, feelings and experiences

1. Open question

2. Closed question

Questionnaire evaluation

1. Cost effective - gather and distribute large amounts of data

2. Reduced effort - researcher doesn’t have to be present

3. Large amount of data quickly

4. Social desirability - answer may be a lie (validity + response bias)

5. Time consuming - long time to design

6. Participant bias

Open questions

1. Does not have a fixed range of answers

2. Respondent is free to answer how they wish

3. Qualitative data

Open question evaluation

1. Increases Detail - expansion on detail

2. New Insight - unexpected answers

3. Useful for sensitive topics - P’s can elaborate

4. Harder to analyse - lots of information

Closed question

1. Fixed number of responses

2. Quantitative data

3. Likert

4. Rating

5. Fixed choice

Rating scale

Participants identify a value that represents their strength of feeling

eg How entertaining do you find zombie films? Circle number

Fixed-choice

A list of possible options an respondents are requires to indicate those which apply to them

eg For what reason do you watch zombie films? Tick all those that apply