Evaluate independent groups
Experimental designs
- participants are not similar to each other, if a mean difference is found this could be due to participant variables
- may act as a cv
- less economical, twice as many p’s needed to produce equivalent data to that collected by repeated measure
- order effects are not a problem
Briefly evaluate repeated measures
Experimental designs
- each p’s have to do at least two tasks and the order may be significant
- to deal with order effects we use counterbalancing= half p’s take part in condition A/B
- could create boredom or fatigue that might cause deterioration in performance on second task
- performance may improve through effects of practice
- demand characteristics are likely
- p variables are controlled and fewer p’s needed
Briefly evaluate matched pairs
Experimental designs
- Order effecs and demand chars not a problem as p’s only take part in one condition
- participants can never be matched exactly
- matching may be time consuming and expensive especially if a pre test is needed
Evaluate random sampling
Sampling
- Unbiased
- C/EV are equally divided between the groups
- Difficult and time consuming to conduct, a complete list of the population may be hard to obtain
- may end up with a sample that is unrepresentative
Evaluate systematic and stratified sampling
Sampling
Systematic= it is objective, once the system for selection is chosen the researcher has no influence over who is chosen, time consuming, may refuse to take part
Stratified= produces representative sample, generalisation possible, identified strata cannot reflect all the ways that people are different so complete representation is not possible
Evaluate opportunity and volunteer sampling
Sampling
Opportunity= convenient, less costly, no need to divide pop into different strata, sample is unrepesentative as it is drawn from a specific area, lacks generalisability, researcher bias as researcher has choice over who is selected
Volunteer= minimal input from researcher, less time consuming, p’s are more engaged, volunteer bias
What is a strength and limitation of lab experiments?
Types of experiments
S= high control over CV and EVs, researcher can ensure that any effect on the DV is likely caused by the manipulation of IV, replication is more possible
L= Lack generalisability, lab env may be artificial and not reflective of natural env, in unfamiliar contexts p’s may act unusually (Low EV), demand chars as the setting of a lab may give it away
What is a strength and limitation of field experiments?
Types of experiments
S= Higher mundane realism than lab because env is more natural, produces behaviour that is more valid and authentic, unaware they are being studied (high EV)
L= Lack of control over CV and EV, effect of IV on DV is hard to establish, replication may not be possible, ethical issues as they do not give consent to be studied
What is a strength and limitation of natural experiments?
Types of experiments
S= provide opportunities for research that may not be undertaken for practical or ethical reasons (e.g romanian orphans), high EV because they involve the study of real world issues
L= naturally occuring events may happen very rarely, limits scope for generalising, p’s may not be randomly allocated to experimental conditions, researcher might be less sure on the effect of IV on DV, may be conducted in a lab so lacks realism
What is a strength and limitation of quasi experiments?
Types of experiments
S= carried out under controlled conditions and share strengths of a lab experiment (replication)
L= Cannot randomly allocate P’s to conditions so there may be CV’s, the IV is not controlled so we cannot claim tat the IV has caused any observed change
What are limitations of all types of observations?
Observational techniques
- Observer bias, observer’s interpretation of a situation may be affected by their expectations, reduced by using more than one observer
- Observational studies cannot demonstrate causal relationships
Evaluate naturalisatic and controlled observations
Observational techniques
Nat= high EV as the findings can be generalised to everyday life as behaviour is studied in normal context
- Lack of control over research situation makes replication difficult
Controlled= may produce findings that cannot be as readily applied to everyday life, CV/EV may be less of a factor so replication becomes easier
Evaluate covert and overt observations
Observational techniques
Covert= removes the problem of demand chars and ensures behaviour is natural (high IV)
- Ethical issues, lack of consent
Overt= more ethically acceptable
- knowledge of being observed may cause p’s to behave unusually
Evaluate participant and non-participant observations
Observational techniques
Participant= researcher can experience the situation as p’s do so they have insight, may increase EV of findings
- may identify too strongly with those they are sutdying and lose objectivity
Non-participant= allow researcher to maintain an objective psychological distance from p’s
- may lose valuable insight as they are too far removed
Evaluate structured and unstructured observations
Observational design
Structured= data produced is likely quantitative which means analysing and comparison is easier, the use of behavioural cat makes recording data easier and systematic
Unstructured= qualitative data, hard to analyse, benefit from richness and depth of detail, may be greater risk of observer bias as objective behavioural categories have not been established
Evaluate behavioural categories
Observational design
- makes data collection structured and objective
- important to make categories as unambiguous and clear as possible
- dustbin categories= where many different behaviours are deposited, categories need to be separate
- should not overlap
Evaluate sampling methods
Observational design
Event= useful when target behaviour is infrequent and could be missed if using time sampling, if event is too complex the researcher may overlook important details
Time= reducing the number of observations that have to be made, may be unrepresentative of the observation as a whole
Briefly evaluate questionnaires
Self report techniques
S= cost effective, large amounts of data quickly as they can be distributed to lots of people quickly, completed without researcher being present so less effort, data produced is straightforward and easy to analyse
L= responses given may not be truthful, social desireability bias and demand chars, response bias where p’s respond in a similar way as they might just be completing it quickly
Briefly evluate the different types of interviews
Self report techniques
Structured= replication is easy, format reduces differences between interviewers, not possible to deviate from the topic which limits richness of data
Unstructured= more flexbility, interviewer can follow up points to gain insight, interviewer bias, analysis is not straightforwards, a risk that interviewees lie for reasons of desirability
What are strengths of correlations?
Correlations
- Useful preliminary tool for research
- They provide a precise and quantifable measure of how two variables are related
- May suggest ideas for possible future research
- Quick and economical, can use secondary data, no need for controlled env
What are limitations of correlations?
Correlations
- The lack of manipulation of varibales and control means that they can only tell us how variables are related, not why
- cannot demonstrate cause and effect, do not know which co variable is causing the other to change
- Intervening variable may be causing the relationship between covariables
- can be misused or misinterpreted, can be presented as causal when they aren’t
Evaluate the use of qualitative data
Types of data
- offers richness of detail
- much broader scopes and gives respondent opportunity to fully report thoughts/ feelings
- greater external validity, provides researcher with more insight
- difficult to analyse
- relies of subjective interpretations to draw conclusions
Evaluate the use of quantitative data
Types of data
- simple to analyse
- comparisons are easily drawn
- less objective and open to bias
- narrower in meaning and detail, may fail to represent real life
Evaluate primary and secondary data
Types of data
Primary= it fits the job, is authentic data obtained from p’s themselves, requires time and effort and resources
Secondary= inexpensive, easily accessed, when examining secondary data the researcher may find desired info is already out there, substantial variation in the quality and accuracy of data, may be outdated or incomplete
-
Exam techniques2
-
Attachment AO181
-
Attachment AO349
-
Memory AO157
-
Memory AO333
-
Approaches in psychology AO170
-
Approaches in psychology AO326
-
Psychopathology AO148
-
Psychopathology AO340
-
Social influence AO152
-
Social influence AO345
-
Research methods AO158
-
Research methods AO324
-
Research methods Y257
-
Gender AO171
-
Gender AO343
-
Forensics AO192
-
Forensics AO356
-
Schizophrenia AO152
-
Schizophrenia AO341
-
Biopsychology AO150
-
Biopsychology AO320
-
Issues and debates AO158
