3.2.3.1 Scientific processes Flashcards
What is an aim?
General statement of what the researcher intends to investigate
What should you always start aims with?
Always start with “TO INVESTIGATE”
What is a hypothesis?
Clear, precise, testable statement that states the relationship between the variables being investigated
When should you write a directional hypothesis?
Used when findings of previous research suggests a particular outcome
When should you write a non-directional hypothesis?
Used when there’s no previous research/previous research is contradictory
State the one-tailed formula (directional)
There will be a significant (state change i.e. increase/decrease) in (DV) for (IV 1 - one which change relates to) than (IV 2)
There will be significant increase in spelling scores out of 20 for children that eat 10g of oily fish a day than children that do not eat any oily fish a day
State the two-tailed formula (non-directional)
There will be a significant difference in (DV) between (IV 1) and (IV 2)
There will be significant difference in spelling scores out of 20 between children eat 10g of oily fish a day and children that do not eat any oily fish a day
State the null hypothesis formula
There will be no significant difference in (DV) between (IV 1) and (IV 2)
There will be no significant difference in spelling scores out of 20 for children between eat 10g of oily fish a day and children that do not eat any oily fish a day
What are variables?
Factors that can change within an investigation
Define Independent Variable
Variable that’s changed/manipulated by researcher
Define Dependent Variable
Variable that’s measured by researcher
Define Extraneous Variable
Variable that may have an effect one results of experiment BUT is not the IV
Define Operationalisation
Clearly defining variables in terms of how they can be measured
Define Standard Procedures
Way of ensuring that all participants receive same instructions & completes same tasks to have consistence in experiment
Why can cause and effect can be achieved in lab experiments?
By having one thing different = researchers can conclude that any differences in measured behaviour of participants (DV) is due to one that differs (IV)
Describe Counterbalancing
To control order effects: participants are split in half
- 1 half completes task 2 followed by task 1
- 1 half completes task 1 followed by task 2
What are order effects?
How order when participants do conditions affects the results (better or bored)
Name 2 types of extraneous variables
- Random Errors
- Constant Errors
What are random errors?
Variables that cannot be predicted
Give an example of random error
e.g. participants state of mind/mood, whether they’re ill/cold on the day
Name 2 ways we can deal with random errors
- Eliminate them by standardising = make sure all participants have same experience
- e.g. same room, same questions, same instructions
- Dealt by randomly allocating participants to experiment and control conditions = effects of errors might be balanced out
What are constant errors?
EVs that have more of an effect on one condition of experiment than other
Name an example of constant errors
Participant characteristics (e.g. personality, intelligence, gender, age)
When do constant errors occur?
When psychologists hasn’t randomised or counterbalance correctly
Participants need to by randomised in ways such as… (name 3 ways)
- Random allocation of order of the conditions
- Random allocation of participants to conditions
- Randomly allocate information in conditions
Name 5 sampling techniques
- Random Sample
- Systematic Sample
- Stratified Sampling
- Opportunity Sample
- Volunteer Sample
Define Random Sampling
All members of target population have equal chance of being selected
Random Sampling
Procedure
List of all members of target population is obtained & names on list are assigned a number
Random Sampling
Name a pro
- Avoids researcher bias
- No influence over who’s selected ∴ researcher can’t choose people that may support their hypothesis
Random Sampling
Name 3 cons
- Difficult and time consuming
- Can still end up with biased sample
- Participants may refuse to take part = biased sample & unrepresentative
Define Systematic Sampling
Every nth member of target population is selected
e.g. every 3rd house on street
Systematic Sampling
Procedure
- Achieved by having sampling frame (list of people in target population in alphabetical order)
- Then sampling system is chosen (every 5th person) or produced randomly to reduce bias
Systematic Sampling
Name 2 pros
- Avoids researcher bias
- Fairly representative
Systematic Sampling
Name a con
Participants may refuse to take part = biased sample & unrepresentative
Define Stratified Sampling
Composition of sample represents the proportions of people in certain sub-groups (strata) within target population
Stratified Sampling
Procedure
- Identify different strata that make up population
- Calculate proportions needed for sample to be representative
- Participants that make up each stratum are selected using random sampling
Stratified Sampling
Name 2 pros
- Avoids research bias
- Representative sample for strata = generalisations possible
Stratified Sampling
Name 2 cons
- Identified strata can’t reflect all ways that people are different
- ≠ Complete representation of target population
- Expensive and time-consuming
Define Opportunity Sample
Involves asking whoever is available at time and is willing to participate
Opportunity Sample
Name a pro
Convenient - saves researcher time/effort & less costly
Opportunity Sample
Name 2 cons
- Unrepresentative sample of target population
- ∵ drawn from very specific area (e.g. 1 street)
- Research has complete control over selection of participants = research bias
Define Volunteer Sampling
Involves participants selecting themselves to be part of sample (self selection)
Volunteer Sampling
Procedure
Get volunteer sample by placing ad in newspaper where target population would see it
Volunteer Sampling
Name a pro
Minimal effort, less time-consuming and costly
Volunteer Sampling
Name a con
Volunteer bias
Asking for volunteers can attract a certain ‘profile’ of person (one who’s helpful, curious)
What are pilot studies?
- Small scale trial run of actual investigation
- Helps to identify any flaws & things that don’t work
Name 3 things pilot studies can be used for
- Experimental studies
- Self report methods (e.g. questionnaires and interviews)
- Observational studies
Give 3 examples of what pilot studies could check for
- Whether standardised instructions are clear
- Check pictures/materials used are clear
- Check participants are given enough time to complete task
- Make sure participants don’t guess the aim
- Make sure participants don’t get bored ∵ there’s too little tasks
What are experimental designs?
Way which participants used in an experiment
Name 3 experimental designs
- Independent Group Design
- Repeated Measures
- Matched Pairs
Describe Independent Group Design
- 2 separate groups of participants experience 2 different conditions of experiment
- e.g. Group 1: participants drink energy drink & Group 2: participants drink water
- Participants only experience one of conditions & performance of 2 groups can be compared
Independent Group Design
Name a pro
- Order effects are not problem
- Participants only take part in one condition = less likely to guess aim, become bored or better at completing the condition
Independent Group Design
Name 2 cons
- Less economical
- 2x as many participants needed
- Individual differences
- Can’t be sure that DV is result of IV = may have been caused by individual differences
- ∴ researcher needs to randomly allocate participants = ensure participants have equal change of being in each group
Describe Repeated Measures
All participants experience both conditions in experiment
Repeated Measures
Name 2 pros
- Individual differences are not an issue
- Fewer participants = cost effective
Repeated Measures
Name 2 cons
- Order effects
- Participants may get bored or better = affects DV or participants’ performance may improve due to practice
- Order of conditions = confounding variable
- ∴ conditions need to be counterbalanced
- Demand characteristics
- More likely to guess the aim and change their behaviour
Describe Matched Pairs
- Participants are matched on some variables that affect DV (e.g. IQ, age)
- In attempt to control confounding variable of participant variables
Matched Pairs
Name a pro
Order effects and demand characteristics less of a problem = do only one condition
Matched Pairs
Name 2 cons
- Participants can never be matched exactly
- Matching = time-consuming and expensive
What are confounding variables?
Variables that affects the dependent variable
When do confounding variables occur?
When extraneous variable remains uncontrolled
e.g. not randomly allocating stimuli or not counterbalancing = become confounding variable
What is the issue with confounding variables?
If you have one = impossible to say whether DV is result of IV or could been a constant error
Describe the British Psychological Society Guidelines
DRIPP
- Deception
- Right to Withdraw
- Informed Consent
- Protection from (psychological) Harm
- Privacy
British Psychological Society Guidelines
Describe Deception
- Deliberately misleading participants
- Make sure debrief
British Psychological Society Guidelines
Describe Right to Withdraw
- All participants should know the withdraw at any time & how to withdraw
- Shouldn’t be be obliged to continue participating and should be aware they leave part way
- Data will be removed
British Psychological Society Guidelines
Describe Informed Consent
Participants should be aware about of aims, procedure, their rights and what their data will be used for
How does the BPS suggest to ask for informed consent?
Participants should get consent letter
State what you should include in a consent letter
Name 4 ways to gain consent
- Parental consent
- Presumptive consent
- Prior general consent
- Retrospective consent
Describe parental consent
If participants under ago of 16, parents will agree on child’s behalf
Describe presumptive consent
- Getting consent from similar group of people (to participants) if study is acceptable
- If group agrees, consent of original participants is ‘presumed’
Describe prior general consent
Participants give their permission to take part in number of studies
Describe retrospective consent
Ask for consent (during debriefing) after taken part in study
Describe protection from (psychological) harm
- Participant should leave in same state as they entered experiment
- Shouldn’t be subjected to physical or psychological harm
- Should be allowed to withdraw if they feel uncomfortable
Describe privacy
- Confidentiality = legal right for participants
- Should ensure anonymity by removing names/identifying details
- Can’t observe people where there’s expectation of privacy
- Data protection
- Right to have personal data protected
- Right to control information about themselves
- If not possible = confidentiality
What is peer review?
- Peer reviewers judge the scientific quality of research
- Process used to help ensure integrity of published scientific work
Why are papers peer reviewed? (name 4 reasons)
- Research that’s published should be of high quality
- Helps allocate research funding
- Helps spot fraudulent research, so research that’s published can be trusted by general public
- To publish of research in science journal and books (must be peer reviewed before hand)
- Contributes to research rating of university departments
Describe the peer review process
- Expert in same field will selected, who works for the journal the paper has been submitted to
- They’ll read unpublished paper & look for mistakes
- Say it can be published, that it needs a review before publication or that it’s rejected
- Usually there are small changes to be made
What are demand characteristics?
Cues that make participants consciously aware of aims of study
What are investigator effects?
- Cues (other than IV) from investigator that encourage certain behaviours in participant
- Might led to fulfilment of investigator’s expectations
- (Cues would be extraneous or confounding variables)
When do investigator effects occur?
When researcher unintentionally or unconsciously influences outcome of research they’re conducting
Investigator effects can be ____ or ______
direct or indirect
How do direct investigator effects come about?
When investigator directly interacts with participants
How do indirect investigator effects come about?
Way investigator has designed study
Name 3 ways investigator effects can occur
- Non-verbal communication
- Physical characteristics
- Bias in interpretation of data
Describe how non-verbal communication can lead to investigator effects
- Researcher can communicate their feelings about what they’re observing without realising
- Participant may then alter their response, affecting validity of data
Describe how physical characteristics can lead to investigator effects
- Appearance of researcher e.g. gender will influence behavioural response of participant
- ∴ behaviour is product of researcher & ∴ not reliable/valid
Investigator Effects
Describe bias in interpretation of data
- May not realise they’re interpreting data in biased way ∵ believe their view is the correct one
- Occurs only for some types of data
Name 3 ways to deal with problems (e.g. investigator effects, demand characteristics) in experiments
- Single blind design
- Double blind design
- Experimental realism
Describe a single blind design
- Participants not aware of research aims/condition their receiving
- Prevents participant from seeking cues about aims and reacting to them
Describe a double blind design
- Both participants and person conducting experiment are blind to aims
- Person conducting investigating = less likely to produce cues of they expect
Describe experimental realism
If researcher makes experiment task sufficiently engaging then participants pays attention to task & not fact they’re being observed
What is validity?
- Means accuracy
- How legitimate something is an explanation of behaviour
Name and describe a test for validity
- Concurrent Validity
- Measures test against benchmark test and high correlation = test has strong validity
What is reliability?
- Consistency
- Repeating test and gaining same result with same group of participants at some point in future
Name 3 types of reliability
- Test-re-Test reliability
- Split Half reliability
- Inter-observer reliability
Describe Test-re-Test reliability
- Assess degree to which test scores are consistent from 1 test administration to the next
- Measurements gathered from single rater who uses same methods or instruments and same testing conditions
- If test is reliable = score would be same on both occasions
