Please enable JavaScript.

Coggle requires JavaScript to display documents.

Scientific processes (aims and hypotheses (stating aims (Precise…

- - - - the group of people that researcher is interested in
      - the group of people from whom a sample is drawn
      - the group about whom generalisations can be made
    - - the group of people, drawn from a population to participate in an experiment
  - - - each member of the population has an equal chance of being selected
      - e.g. selecting names out of a hat
    - - unbiased selection
        
        increases chances of getting unbiased + representative sample
      - generalisation
        
        results can be generalised to target population because it is representative
    - - not always unbiased
        
        sample not always representative
        
        e.g. could be all females
      - need details of population
        
        need full list of population
        
        not all of those selected wish to take part
  - - - Taking every nth person from a list
    - - calculate population size
      - figure out how many you need in the sample
      - divide population size by sample size to get nth number
    - - unbiased selection
        
        increased chance of having unbiased sample that is more representative
      - if representative it is more generaliseable
    - - periodic traits
        
        sampling interval can interact with a periodic trait of the same interval meaning that get a certain type of person
      - Not always unbiased sample
        
        e.g. could be all females
  - - - small scale reproduction of the population
    - - representative
        
        subgroups = representative of population
      - unbiased because of the random sampling in the sub-groups
    - - divide population into characteristics important for the research
      - see what proportion of population have those characteristics
      - have same proportion of characteristics in sample e.g. if 12% of population are in between 20-30 years that will be reflected in the sample
    - - knowledge of population required
        
        details about the population characteristics may not be available
      - Time consuming
  - - - selecting participants that are available and willing to take part
      - e.g. asking people in a street
    - - ease of formation
        
        using people who are readily available
        
        Has to be used in natural experiments as researcher has no control over who is being studied
    - - Unrepresentative
        
        likely to be biased because uses people readily available and these are likely to be a certain type e.g. those not in school
      - Self selection
        
        a certain type of people are going to be willing to take part so not representative (demand characteristics?)
  - - - people volunteer to participate
      - e.g. reply to an advert
    - - Unrepresentative
        
        people willing to participate tend to be a certain type so can't generalise to target population
      - Demand characteristics
        
        Those willing to participate are eager to please so increased chance of demand characteristics
    - - ease of formation
        
        requires little effort from researcher
      - Less chance of "screw you" phenonmenon
        
        eager to participate so not going to sabotage results
- - - - means that independent observers can code in the same way
      - lessens the chance of observer bias
  - - - number of times behaviour occurs in target individuals
    - - counting/describing behaviour in a set time frame
  - - - observers become actively involved in situation they are studying
    - - researchers do not become involved in situation they are studying
    - - participants are aware they are being observed
    - - participants are not aware they are being observed
- - - - filter reducing chance of flawed, unscientific research being accepted as a fact
    - - Allocation of research funding
        
        public bodies need reviews to help them decide which research his likely to be worthwhile
      - Publication of research in academic journals and books
        
        prevents faulty research entering public domain
      - assessing the research rating of university departments
        
        future funding for these departments depend on receiving good ratings from peer reviews
    - - new solutions needed to maintain quality of information when volume and pace of information available on the internet is so large
      - readers often decide whether this info is valid or not
        
        this means "peer" is coming to mean everyone
  - - - then send back improvement suggestions and problems
      - either recommend to
        
        accept work unconditionally
        
        accept it on conditions of improvement
        
        reject it but able to resubmit
        
        reject it outright
    - - reviewer anonymity
      - means unbiased because there is no influence by the researcher
      - BUT reviewer could delay process and publish research themselves or be unnecessarily harsh
    - - bias based on ethnicity, gender etc. wont happen
      - BUT researcher could be identifiable by writing/research style
    - - reduces risk of personal comments + plagiarism
      - BUT may be less harsh
      - encourages honesty and openness
  - - - narrow social world of research = relationship always come in to play
      - reviewers may be influenced by being sponsored by certain companies that want certain research published
- - - - no order effects
      - less demand characteristics
      - participants being matched reduces individual differences
    - - need more participants
      - matching is difficult and time consuming
  - - - no individual differences
      - produces more data with fewer participants = cheaper
    - - order effects
      - doing all conditions means increased chance of demand characteristics
      - takes more time (especially with gap to counteract boredom)
  - - - using different participants for each condition means there are no order effects
      - only doing one condition means they are less likely to guess the research aim reducing demand characteristics
      - both sets of participants can be tested and the same time which saves time and money
    - - more participants are needed
      - individual differences means that change in DV may be due to this instead of manipulation of IV
- - - - structured
        
        closed questions with interviewers writing down answers
      - unstructured
        
        informal discussion on particular topic
        
        training needed
      - semi-structured - combining structured and unstructured to produce quantitative and qualitative data
      - need to decide interviewer
        
        interpersonal factors such as gender, ethnicity, age may affect participant answers
      - different roles and language can affect the interview
      - training needed
        
        need to listen and understand
        
        need to relax participants so non-verbal communication is important
        
        difficult questions should go at the end when pps are more relaxed
    - - better for complex issues because interviewer can clear up misunderstandings
      - sensitive issues can be discussed because interviewers can make people feel relaxed
      - semi structured provides both quantitative and qualitative data analysis
      - more standardised interviews can be replicated
    - - interviewer effects such as gender of interviewer can mean participants give biased answers
      - interviewer training needed
      - ethical issues
        
        may not know purpose of interview
        
        may reveal more than they wished to
      - not suited to people who find it difficult to express their thoughts verbally
  - - - Open questions
        
        Participants answer in their own words
        
        difficult to analyse but allow freedom of expression
      - Closed questions
        
        usually yes/no questions
        
        easy to quantify but restrict answers
        
        can only have certain answers
    - - Quick
        
        large amounts of info can be collected in short amount of time
      - cheap
        
        can send out questionnaires cheaply
      - lack of investigator effects
        
        completed without researcher present
      - can lead to both quantitative and qualitative analysis
      - If used standardised questions, can be replicated
    - - misunderstanding
        
        pps could misinterpret questions
        
        leading questions
        
        emotional language
      - biased samples
        
        volunteer samples often used
      - low response rates
        
        means are uneconomical
      - not suitable for sensitive issues
      - social desirability
        
        may lie and give idealised answers
    - - aim
        
        exact aim makes it easier to write questions
      - length
        
        if short pps more likely to complete it
      - questions
        
        concise, unambiguous, easy to understand
      - measurement scale
        
        used to assess things like attitude levels
- - - - directly manipulated to test its effects on the DV
    - - measurable outcome of the action of the independent variable in an experiment
    - - need to be controlled so that they do not vary
      - participant variables
        
        age, gender, intelligence etc.
      - situational variables
        
        surrounding environment during experiment such as noise
        
        experimenter variables
        
        e.g. change in personality appearance of researcher (male/female)
    - - extraneous variables that can affect the DV an confuse results
      - kind of replacing IV
  - - - may mean specifying way they are going to be measured
- - - - guessing purpose of research and trying to get right results to please investigator
      - trying to get wrong results = "screw you effect"
      - acting unnaturally out of nervousness of evaluation
      - NOT social desirability bias
    - - single blind procedure: participants do not know what condition of study they are in
      - experimental realism: making task engaging enough so pps don't think about being observed
  - - - investigators unconsciously encourage pps e.g. Rosenthal: males are more pleasant and friendly with female pps than male ones
      - could ask question that leads participants to answer in a certain way
      - Indirect investigator effects
        
        may operationalise measurement of variables so that desired result is more likely
        
        "investigator loose procedure effect": investigator doesn't clearly specify standardised instructions which leaves room for result to be influenced by experimenter
      - how to avoid it
        
        double blind procedure: both participants and investigator don't know which condition they are in
        
        participant and person conducting investigation don't know aims
        
        means person conducting it is less likely to produce cues
- - - - Psychologist should tell participants true aims of the study or at least what is going to happen
      - Might lead to participants guessing aims of study and demand characteristics
      - Participants must know what is going to happen so they can make an informed decision on whether to take part in the study or not
      - must also know risks and benefits of study
      - those under 16 need parental consent and those under the influence can't participate
    - - pps shouldn't be misled without justification
      - unacceptable is participants are likely to be uncomfortable in debriefing
      - means informed consent can't be made
    - - Participants must be aware of this
      - They must be able to withdraw their data afterwards as well
    - - health and dignity of the participants should be considered
      - If during research there are negative consequences it should be stopped immediately and they should try to rectify it
      - risk of harm should be no greater than in ordinary life
    - - confidentiality means data can be tracked back to names (means that they can follow up)
      - names shouldn't be disclosed unless agreed in advance
      - anonymity: data can't be tracked back to names (numbers applied instead)
      - important when researcher publishes findings
    - - researcher should consult peer, likely participants and consider alternative methods
      - should do cost-benefit analysis
      - Researcher must assume responsibility
      - if colleague has issue with research they should talk to the researcher
  - - - able to rectify harm and deal with any concerns participants have