Psych: Research Methods

• Experimental Methods = attempt to find a cause and effect relationship between IV & DV, and measure the extent of this effect.

  include:

  1. Lab
  2. Field
  3. Natural
  4. Quasi
• LAB EXPERIMENTS:

  • Controlled conditions - to minimise effect of extraneous v's
  • Manipulate IV to measure the effect on DV
  • Participants aware they're taking part in investigation - due to the contrived nature of the situation, unlike real life
• EVALUATION LAB EXPERIMENTS:

  • :) High degree of control over extraneous v's, prevent them from becoming confounding v's and negatively affecting DV. Provides high degree of internal validity, allowing conclusions to be drawn
  • :( Lack external validity. Artificial nature of the environment means the study may lack ecological validity. Means that the findings cannot always be generalised to settings beyond the lab as the tasks lack mundane realism and wouldn't be everyday occurrences. As ptps aware of the study, may behave in an unnatural manner resulting in demand characteristics.
• FIELD EXPERIMENTS:

  • Natural conditions - researcher manipulates IV to measure affect on DV
  • 'field' is considered any location other than a lab
  • ptps typically don't know they're taking part in the investigation, observing more natural behaviour
• EVALUATION OF FIELD EXPERIMENTS:

  • :) Higher level of ecological validity in comparison to lab. Results more likely to be representative of behaviour witnessed in everyday life --> HOWEVER there's less control over extraneous v's which can become confounding v's and distort the findings
  • :( Ethical Issues. Ptps unaware of participation and so cannot give informed consent to take part. As such, research may breach their privacy rights and a cost benefit analysis will need to be conducted before proceeding with any study to ensure the perceived outcomes will outweigh any personal costs
• NATURAL EXPERIMENT:

  • No manipulation of IV - instead examines effect of an existing IV on the DV
  • The IV = naturally occurring [i.e. flood/earthquake] and behaviour of people affected is either compared to their behaviour beforehand or with a control group who haven't encountered IV
  • Ptps could be tested in a lab as a part of the study
• EVALUATION OF NATURAL EXPERIMENT:

  :) Naturally occurring IV means a higher level of external validity compared to lab/field experiments. Considered high in ecological validity as real-life issues being studied rather than manipulated.

  :)Unique insights gained into real life situations --> HOWEVER Naturally occurring events, that interest the researcher, may only occur very rarely - limits the opportunity to generalise the results to similar events or circumstances

  :( No control over the environment & subsequent extraneous v's means it's difficult to accurately assess the effects of IV on DV.
• QUASI EXPERIMENTS:

  • Naturally occurring IV, that already exists
  • IV = difference between people [i.e. age, gender, personality trait]
  • Often carried out in natural settings, but don't have to be
• Evaluating Quasi Experiments

  • Participants cannot be randomly allocated to conditions (to remove the issue of bias)
  • Independent variable is naturally occurring and so the level of IV to which the participants belong is predecided
  • Less certain that the IV alone will have caused the effect - other dispositional/environmental factors may influence
• Evaluating Quasi Experiments

  • Allow researchers to compare different types of people easily to provide insight into similarities/differences
• Evaluating Quasi Experiments

  • Methodological issues associated
  • No control over environment and subsequent extraneous variables (when in a natural environment)
  • When done in lab conditions, the high level of control means research lacks ecological validity, and findings can't always be generalised to real life settings - behaviour may not always translate
• OBSERVATIONAL TECHNIQUES:

  * Covert and Overt

  * Participant and Non participant

  * Naturalistic and Controlled

  * Structured and Unstructured

  
  

  --> these techniques aren't mutually exclusive
• COVERT OBSERVATIONS:

  • 'undisclosed' observation - consists of observing people without their knowledge
  • May be made aware of their involvement after the observation has taken place
• Evaluating Covert Observations


  :) Investigator effects less likely. Investigator is hidden so less chance their direct/indirect behaviour will have an impact on the performance of the ptps. Consequently, less chance of demand characteristics as unaware of observation. Behaviour will be more natural and representative of everyday behaviour

  :( Ethical Issues - ptps not aware, cannot give fully informed consent or exercise their right to withdraw

  --> Can observe in a public place [such as a shopping centre] as long as assessment is made to ensure no privacy laws are being violated
• OVERT OBSERVATION:

  • Observations are 'open' and ptps aware they're being observed e.g. filming publicly [overt, non-ptp] or joining a class and informing other students you're carrying out an observation [overt participant]
• Evaluating Overt Observations:

  :) More ethical - ptp aware so can inform them in advance of the aims and thus obtain informed consent. Awareness of participation allows them to exercise right to withdraw - reputation of psychological research as being ethical is protected

  :( Possibility of investigator effect. Possible for bias to occur whereby what the investigator does influences behaviour of ptps in a way not intended [e.g. body lang/facial expressions] May change behaviour thru demand characteristics - reducing internal validity
• PARTICIPANT OBSERVATION

  The person conducting the observation also takes part in the activity that's being observed
• Evaluating Participant Observations:

  :) Can obtain in depth data as in close proximity to the ptps can gain a unique insight into the phenomenon of the question. Also unlikely to overlook any behaviour that would be missed due to nuances only seen by becoming a ptp. Comprehensive understanding of human beh acheived

  :( Possibility of investigator effects [concious/unconscious]. consequently, may show demand characteristics -> not observing natural behaviour -> reduce internal validity
• NON-PARTICIPANT OBSERVATION:

  Person conducting the observation doesn't participate in activity being observed. Quite common in educational settings, e.g. sit in the corner of the room and observes.

  Aim for the observer is to be as unintrusive as possible and not engage with any of the activities taking place
• Evaluating Non-Participant Observations:

  :) Investigator effects less likely. Usually observing from a distance - in some situations not at all visible - so unlikely to have a negative impact. Behaviour more likely to be representative of natural and unaltered human conduct

  :( Due to lack of proximity, might miss behaviours of interest, meaning unique insights which contribute to the understanding of human behaviour will be overlooked because of not being personally involved
• NATURALISTIC OBSERVATIONS:

  Carried out in a non altered setting in which observer does not interfere in any way.
• Evaluating Naturalistic Observations:

  :) Higher level of ecological validity can be achieved. Researcher records naturally occurring behaviour in the original environment -> likely to be representative of everyday behaviour, and reflect spontaneous actions that sometimes occur incidentally

  :( Issues of reliability - as they're naturally occuring it can be hard, if not impossible, for the exact same conditions to be replicated. consequently the test-retest method for checking reliability cannot be used as researcher is not in control of the variables - lacks replicability
• CONTROLLED OBSERVATIONS:

  Conducted under strict conditions where extraneous variables can be controlled. Sometimes one way mirrors can be used for these types of observations.
• Evaluating Controlled Observations:

  :) Can be replicated to check for reliability. Means that standardised procedures, manipulation of IV and control over extraneous v's can be repeated by the same, or different, researchers to assess reliability.

  :( Lower level of external validity. Recording behaviour in artificial environment with variables subject to strict manipulation -> setting of the observation feels unnatural -> in response behaviour may alter -> the observation no longer represents real life occurrences -> ecological validity is then questioned
• STRUCTURED OBSERVATION:

  Researcher uses coded schedules according to previously agreed formula to document the behaviour and organise the behaviour into behavioural categories. behavioural category is when psychologists must decide which specific behaviours should be examined -> involves breaking the target behaviour into measurable and observable components
• Evaluating Structured Observations:

  :) Researcher can compare behaviour between participants and across groups. Use of operationalised behavioural categories makes coding of data more systematic. Standardised behaviour schedule results in greater inter-rater reliability

  :( Internal validity. Researcher may miss some crucial behaviours -> may not provide the full picture about behaviour -> lack finer details. Limited scope
• UNSTRUCTURED OBSERVATION:

  Every instance of the observed behaviour recorded amd described in as much detail as possible
• Evaluating Unstructured Observations:

  :) Rich data obtained, able to obtain a comprehensive view of human behaviour. Adds to the internal validity of observational technique

  :( Prone to observer bias due to lack of objective behaviour categories

  -> may only record behaviour which is of subjective value to them, and not a valid representation of what is being displayed -> as a result there may be issues with inter-observer reliability -> lacking consistency in observations recorded.