Data collection techniques

Research Design

• Experimental Design- The indepedent variable(s) manipulated and the depedent variable(s) measured.Causal relationships are clear
• Lab experiments- Carried out in lab with high levels of control
• - High control to prevent confounding variables, Standardised procedure makes replication easy
• - Artificial environment may change behaviour, Lacks ecological validity, Demand characteristics
• Field experiments- Carried out in environment but variables still manipulated
• - More realistic so higher ecological validity (natural behaviour). Less demand characteristics
• - Less control over confounding variables, reliability affected
• Natural Experiments- carried out in environment but variables occur by chance
• - High ecological, low demand characteristics, use where ethical restrictions prevent research.
• - Costly, no control over extraneous variables, so reliability affected
• Correlational Design-
• All variables are measured and the association between them is assessed. Might be able to draw causal relationships through theory or other considerations. Naturalistic observation and surveys often used.
• - Can be used where ethical restrictions prevent research, Can help to provide confirmatory evidence, it allows researchers to predict strength and direction of relationship.
• - You cant establish causation unless a logical sequence of relationship exists.

Data Collection methods

• Observational methods
• - Behaviour observed where it naturally occurs, participants may be aware (overt) or unaware (covert). 3 main types: controlled observation, naturalistic observation, participant observation
• Case Studies
• - Individual group/event etc is examined in detail, multiple techniques may be used. Often uses idiographic approach, often used in clinical psychology and psychiatry.
• Surveys
• - Questions given to sample drawn from larger population, relies on self-report. Types: in-person, telephone, mail-out, web-based
• Interviews
• - smaller sample give rich info in 'live' interaction with researcher (meaning in person or on telephone)
• Experiments
• - IV's manipulated and DV measured to measure effect, can be lab or field based
• New Technologies
• - Allows new data collection, using internet, smartphone apps and ambulatory data monitoring. Also VR and social media

Observational Studies- Controlled

• Usually done in lab environment for high control.
• Behaviour is usually systematically classified and coded into distinct categories.
• Behaviour may be coded by more than one researcher (inter-rater reliability)
• Coding
• The coding may use numbers or a scale to measure behaviour
• Key examples
• - Bandura - Bobo Doll Studies 1961
• - Ainsworth- Strange Situation 1970
• Strengths
• - Easy to replicate using same methods and observation schedule
• - Data quick to analyse using quantitative methods
• - Relatively quick to conduct, so large samples can be used
• Limitations
• - Hawthorne effect/ Demand characteristics may affect validity
• - Participants may act differently if they know they are being observed (overt)

Observational Design- Naturalistic

• - Behaviour is observed in its natural environment
• - May use a host of methods to record data
• Coding
• May involve numbers or a scale to measure behaviour
• Key Examples
• - Croft- Killer whale menopause (2017)
• - Levine- Group violence (2011)
• Strengths
• - Observing behaviour in its natural setting increases ecological validity
• - This type of research is often used to inform further research
• Limitations
• - Observation on a small scale may not be representative
• - Difficult to replicate
• - Substantial training needed- how to code, what to observe, using categories etc.
• - may not be able to draw cause and effect or direction

Observational Design- Participant

• - A variation of naturalistic observation
• - researcher goes 'undercover' as a participant to conduct covert observation
• Example
• Festinger- cults and cognitive dissonance (1957)
• Strengths
• - Higher ecological validity
• - As participants dont know they are being observed by another participant, their behaviour wont chnge or be affected by demand characteristics
• Limitations
• - There is difficulty recording the data as the researcher must remain in character
• - Risk of a loss of objectivity and researcher bias

Observational- recording data

• Time- Sampling
• The events being measured are agreed in advance and the coding established. Observation happens within a specific period or on a schedule. At all other times, behaviour is ignored
• Event- Sampling
• Events are agreed in advance and the coding established. Event frequency and other characteristics get recorded. Other behaviours are ignored.
• Instantaneous- Sampling
• Observations made at a specific time-point. All observations before or after this are ignored

Case Studies

• Individual group/event investigated in detail, providing with rich data. Multiple techniques may be used (survey, observation etc). Often uses the idiographic approach:
• - observe what happens to a participant, and reconstructing the 'case history' of a person/group/event etc
• Often used in clinical psychology and psychiatry
• - it provides a rich source of information and insight for further research
• Strengths:
• - Offers rich descriptive information
• - Often indicated hypotheses for futher research
• - Gives opportunity to study rare phenomena in detail
• - Allows you to investigate factors that couldnt be ethically experimented on
• Limitations:
• - hard to establish cause and effect (causal interference)
• - Not generaliseable or representative
• - Relies on subjective interpretation of data
• - Very hard to replicate

Surveys

• Types of Surveys
• In person Surveys- Person completes it in presence of researcher
• - Control over environment reduces biases
• - researcher can provide explicit instructions
• - highest quality survey data
• - Very time consuming and costly
• - Small samples because of the expense
• Telephone - Researcher phones participant
• - Cheaper than in-person so larger samples are possible
• - anonymous
• - less control so less credibility
• - visual materials unavailable and materials are less sophisticated
• Mail-out- survey is sent to a collection of addresses
• - Relatively cheap and convenient
• - Participants have enough time to complete it
• - no interviewer bias
• - visual aids provided
• - Rapid data collection
• - Anonymous
• - Self-selection bias
• - full instructions required
• Web based- Survey available online, researcher invited ppts
• - Cheap, convenient, participants have time to complete
• - no interviewer bias, rapid data, anonymous and has visual aids
• - self selection bias and full instructions required

Interviews

• Strengths
• - representative sample provides rich in detail information about the population of interest.
• Limitations
• - if sample is unrepresentative may lead to false conclusions,
• - it is also open to researcher and social desirability biases
• Open vs Closed Questions
• - open questions give more rich data but are hard to analyse using empirical methods- instead must use qualitative
• - Closed questions give more specific answers but are more open to bias from the wording of the question

Experiments

• Main Features
• IV manipulated (the cause) and the DV is measured (the effect). They can take place in field or lab environments.
• Strengths:
• - best method if you want to establish cause and effect relationships
• - you can phsycially control extraneous or confounding variables- this leads to higher validity
• Limitations
• - Faulty assumptions of validity and reliability in the findings may lead to drawing false conclusions.

Internet/ Digital therapy

• - This allows for the delivery of psychological treatments via computers, the internet, mobile devices or apps
• - These are being developed for a wide range of psychological problems, such as depression, anxiety or insomnia. Also for transdiagnostic interventions
• - most of these interventions deliver packages or CBT but increasingly they are using "third wave" or other psychotherapy e.g: ACT, IPT, mindfulness and problem-solving therapy
• - the therapies can be a linear program of sessions, or individually chosen modules
• - often includes a combination of the classic session content, but with interactive tools, self monitoring tools, audio and visual files.
• Review- Fairburn and Patel 2017
• - popular and can access under-served groups
• - Online clinics can obtain necessary clinical change on large scale
• - supported interventions more effective then unsupported
• - supported digital-treatments as effective as face-to-face
• Key concerns
• - Lack of research comparing the two treatments that are synonymous (e.g: face to face vs digital treatment for depression)- we cant do that as there is no one treatment established yet
• - Effect of digital delivery- older people may not find it accessible etc
• - How do these treatments work?

Slide 12

Apps and Ambulatory Monitoring

• Ambulatory monitors are small, portable electrocardiograph machines that are able to record the heart's rhythm.
• - Allows research in monitoring of mood, behaviour and physiological responses in a realistic setting
• - Allows for interventions supporting behavioural change, provides real-time feedback and application of newly-acquired skills in real life situations.
• Goal Processes and Mood Disorders
• goals are central to psychotherapeutic work with mood disorders. disruption of goal content, structure and processes have been implicated in mood disorders. These technologies can provide a real life measurement of goal processes
• Exeter's input to this field
• - basic research (ESRC funded)
• - Potential interventions (ADEPT, MOODFOOD)
• Training of Clinical Psychologists
• Inhibition training
• this is 'impulse training'- personalised 'brain training' leads to new associative relationships between cue and response.
• Exeter's input in this field
• - Natalia Lawrence- Obesity and junk food
• - other coleagues- substance abuse, traffic offending

Virtual Reality

• "... a way for humans to visualise, manipulate and interact with computers and extremely complex data."- Austakalnis and Blatner 1992
• - Advanced form of human-computer interface (Rizzo et al., 1997) that allows the user to interact with computers and digital content in a more natural or sophisticated fashion
• Immersive
• Use of digital technologies and body tracking in a stimulated environment which allows embodiment in the situation. (E.g: head mounted devices, 360 degree/3D video)
• Non-immersive/ Semi-immersive
• Modern computer console game system to present graphic environments allowing users to naviagte and interact (E.G: driving simulator, treadmills)
• VR in Social Psychology
• Increases ppts psychological engagement in research, so you can understand behaviours in more naturalistic setting
• VR in Clinical Psychology
• ASSESSMENT: Understand behaviours in more natural setting (ADHD)
• THERAPY: Simulating situations, acquiring and applying new skills in safe environment
• TRAINING: Simulating therapy situations/client responses for giving feedback about therapists’ skills and effectiveness
• Challenges for Development
• - Coding skills are required

Social Media

• Allows access for hard to reach participants
• Can be run by established platforms e.g: Facebook, so trustworthy
• Content like blogs and forums can be mined for quantitative data or be set up for research purposes
• Key Issues
• - informed consent
• - anonymity and confidentiality