research methods y1&2 Flashcards
what is an experimental method?
- manipulation of IV
- so that the IV can have an effect on the DV
- which is measured and stated in results
what are the 4 types of experimental methods?
- quasi
-laboratory
-field
-natural
what is an independent variable
the variable that the experimenter changes / manipulates
- e.g : temperature of the room ( experimenter changes this , to see the change in maths scores )
what is a dependant variable
- variable being tested and measured in an experiment
- it is “dependant” on the independent variable
e.g : measuring the maths scores of participants in different temp conditions
Aim
general statement that the researcher intends to investigate
Hypothesis
A detailed statement which is clear, precise and testable that states the relationship between variables being tested.
Directional hypothesis
The researcher makes it clear what difference is anticipated between the 2 conditions or groups.
Clear effect of iv on dv
(One tailed).
e.g “ “The more sleep a participant has the better their memory performance.”
Non-directional hypothesis
Simply states that there is a difference but not what the difference will be.
e.g : “The difference in the amount of hours of sleep a participant has will have an effect on their memory performance, which will be shown by the difference in the memory test scores of the participants.”
Why must factors that effect the DV be controlled?
- extraneous variables
- confounding variables
- to make sure that the effect on the DV is purely due to the independant variable
How would you test the effect of an IV
Compare the different experimental conditions:
- Control condition (e.g no energy drink/water) = used to determine whether the IV affected the DV.
- Experimental condition (e.g energy drink)
Operationalisation
Clearly defining variables on terms of how they can be measured = makes the hypothesis clear + testable.
Example: After drinking 500ml of energy drink, participants speak more words in the next 5 minutes than participants who drink 500ml of water.
(even more operationalised : number of words said)
Extraneous variables
Any unwanted variables outside of the IV that will impact the DV.
- Researcher should minimise the influence (control) or remove these variables.
e.g : lighting of lab or age of participants
Confounding variable
An uncontrolled extraneous variable that change systematically with the IV and affect the DV, so results won’t show the effect of the intended IV.
e.g : time of day
to control : all participants take test same time of day
state 3 types of extraneous variables
- Participant variables
- Situational variables
- Investigator effects
Outline examples of participant variables
- Personality
- Age
- Intelligence
- Gender
- Participant reactivity
Explain how to control Participant variables
-Sample: Use random sampling to gain a representative sample from the population.
-Design: Use repeated measures or matched pairs
Allocation: Randomly allocate them to conditions
Outline examples of situational variables
- Time of day
- Heat
- Demand characteristics
Explain how to control Situational variables
- Standardise: Keep everything the same for each participant (procedures and instruction)
- Counterbalance: Reduces effect of situational variables
Definition and examples of investigator effects
Subtle cues from a researcher that may affect the performance of participants in studies:
- Body language
- Tone/voice
- Bias
Explain how to control Investigator effects
- Double blind: Neither researcher nor participants knows which condition they’re in.
- Inter-rater: Independent raters rate the same behaviour as the researchers and check for agreements.
Outline the definition of counterbalancing
- participant sample is divided into a half
- one half completing the two conditions in one order
- the other half completing the conditions in the opposite order
- used to deal with order effects e.g when using a repeated measures design
Demand characteristics
A cue that makes participants unconsciously aware of the aims of a study and helps them work out what the researcher expects them to find.
- May behave in an unnatural way and over/under-perform to please the researcher = affects results/DV
How to control demand characteristics
- Deception: Use distractor questions and lie about the aim.
- Single blind: Participant is unaware of which condition they’re in.
What can demand characteristics cause?
Please-U effect : may act in a way they think the researcher wants them
Screw-U effect : intentionally underperform to sabotage the study’s results
Participant reactivity
when the responses and/or behaviours of study participants are affected by their awareness that they are part of a study
can lead to :
- demand characteristics
-investigator effects
Randomisation
used in the presentation of trials in an experiment to avoid any systematic errors that might occurs as a result of the order in which the trials take place.
Standardisation
using the exact same formalised procedures and instructions for every single participant involved in the research process
-eliminates non standardised instructions as being possible extraneous variables
Laboratory experiment
Conducted in a highly controlled environment, where the researcher manipulates the IV and records the effect on the DV.
- Strict control is maintained over extraneous variables
Strength of Laboratory experiment
- High internal validity due to control over extraneous variables = Researcher can ensure any effect on the DV is due to their manipulation of the IV + proves cause and effect. ( high degree of control )
- Results are more replicable = Results are valid and generalisable :
Limitations of Laboratory experiments
- Lacks ecological validity = Not true to real world so can’t be generalised
- Hawthorne effect = behaviour is altered due to awareness of the study.
- Demand characteristics = participants are aware of study due to lab conditions so behaviour is unnatural.
Field experiment
Conducted in a natural environment where the researcher manipulates the IV and records the effect on the DV.
e.g Hoflings hospital study on obedience
Strength of field experiments
- High mundane realism = Environment is more natural, so behaviour is more valid + authentic.
- High external validity = Participants are unaware of study.
Limitations of field experiments
- Lack of informed consent = ethical issues, invasion of privacy.
- Increased realism also increases extraneous variables = Cause and effect of IV + DV is harder to establish and precise replication won’t be possible.
Definition of Natural experiment
Researcher takes advantage of a pre-existing IV.
- Natural as the variable would’ve changed regardless of the researchers interest in it.
e.g : biological explanations of bullying
Strengths of natural experiment
- Provide research opportunities for studies that can’t be conducted due to ethical/practical reasons.
- High external validity = involves study of real situations
Limitation of natural experiment
- Naturally occurring event is rare = limits scope for generalising results to other similar situations.
- Participants may not be allocated randomly to experimental conditions = less clarity that the effect on DV is due to the IV.
Quasi experiment
IV isn’t determined by anyone, but is not manipulated and an existing difference between people. (age , gender, phobias)
e.g A memory task with a group of clinically depressed participants compared to a control group of non-depressed participants
Strength of Quasi experiments
- High internal validity due to control over extraneous variables = Researcher can ensure any effect on the DV is due to their manipulation of the IV + proves cause and effect.
- Conducted under highly controlled conditions = replicable, reliable, generalisable results.
Limitation of quasi experiments
- Can’t randomly allocate participants to conditions = possibility of confounding variables
Population
A group of people from whom samples are drawn
Generalisation
The extent to which research results can be applied to the population.
Sample
A group of people who take part in a research investigation, chosen from a population.
Bias
When certain groups may be over or under-represented within the selected sample
= limits extent of generalisation.
Random sampling
- Obtain a list of the population and assign numbers
- Randomly choose sample via lottery method (random number generator or names in a hat)
Strengths of random sampling
- No researcher bias = unable to choose samples to support their hypothesis.
- Everyone has an equal chance of being chosen
- Laws of probability = likely to be representative
Limitation of random sampling
- Difficult and time consuming to obtain a list of the population.
- Possibility of unrepresentative sample = doesn’t reflect the distribution of characteristics in the population.
- Participants may refuse to participate = becomes a volunteer sample.
Systematic sampling
- Sampling frame is produced = list of population in alphabetical number.
- Sampling system is made = every nth member of the population is chosen.
Strength of systematic sampling
No researcher bias = researcher has no influence over chosen people after the sampling system is chosen.
- Fairly representative
Limitation of systematic sampling
- Complete representation isn’t possible = doesn’t reflect all differences in people.
Stratified sampling
- Identify the different strata within the population + work out the proportion needed from each strata to be representative.
- Use random sampling to choose people from each strata.
Strength of stratified sampling
- No researcher bias = sample from strata is random and uninfluenced by the researcher.
- Representative = designed to accurately reflect composition of the population, so generalisation is possible.
Limitation of stratified sampling
- Complete representation isn’t possible = can’t reflect all the ways people are different.
Opportunity sampling
- Select anyone willing + available by asking anyone around at the time of their study. (e.g the streets)
- participants available at the time of study
Strength of opportunity sampling
- Convenient : easy method of recruitment
- Saves time and effort
- Less costly
Limitation of opportunity sampling
- 2 forms of Bias:
= Researcher bias as they have complete control over the selection.
= Unrepresentative as chosen from a specific area so can’t be generalised.
Volunteer sampling
- Participants select themselves to be apart of the sample (e.g raising hands/advert)
strength of volunteer sampling
- Requires minimal researcher input : willing participants , more likely to co-operate
- Less time-consuming
Limitation of volunteer sampling
Volunteer bias = attracts helpful, curious people so generalisation may be difficult.
- motivation like money could be driving participants so participants may not take study as seriously
independant group design
Two separate groups experience two different conditions of the experiment.
- One group takes part in condition A and the other group takes part in condition B.
- Performances of the groups are compared
Limitation and resolution of independant group design
- Researcher can’t control participant variables = different abilities of participants
Randomly allocate participants to conditions so the participant variables are distributed evenly.
- Needs more participants
Strengths of independent group design
no order effects : e.g practise effect or boredom effec
- Can use the same test for both groups = faster
- Participants are less likely to guess aim
Repeated measures design
Only one group of participants and they take part in both conditions
Limitation and resolutions of repeated measures design
- Order effect may affect performance (participants may perform better/worse in the second condition due to practice/boredom)
= Use 2 different tests to reduce practice effect, so the order effect is dealt with via counterbalancing.
- During the second test, participants may guess the aim of the experiment which affects their behaviour
= Use deception and lie about the aim + use distractor questions
Strength of repeated measures design
- Limits the variability between participants
- Fewer participants needed
Matched pairs design
Two separate groups, but they’re matched into pairs for certain qualities before splitting (age,gender,intelligence).
each person from a pair goes into a different experimental condition
Limitations and resolution of matched pair design
- Very time consuming + difficult to match participants based on key variables = Restrict number of variables to match
- Not possible to control all participant variables = Conduct a pilot study to consider key variables that may be important when matching.
Strengths of matched pair design
- Reduces participant variables as researcher has paired the participants, so each condition has people with similar abilities + characteristics.
- Avoids order effects, so no counterbalancing is needed.
demand characteristics less of a problem
hawthorne effect
When an individual modifies an aspect of their behaviour, due to their awareness of being observed.
Mudane realism
The extent to which the materials + procedures involved in a study are similar to events that occur in the real world.
External validity
Extent of results being applicable to other experiments, settings, people, and times.
Naturalistic observation
Watching and recording behaviour in a setting where it would normally occur.
Strength of naturalistic observation
- High ecological validity = results can be generalised to everyday life as behaviour is observed in a natural env
- No researcher influence
- People are less likely to alter their behaviour as they’re unaware of the observation + it’s not controlled
Limitation of naturalistic obervation
- Low internal validity = no control over extraneous variables, so it confounds + difficult to judge behaviour patterns.
- Hard to replicate = due to no control and extraneous variables.
Controlled obervation
Watching and recording behaviour within a structured environment, where variables are controlled
Strengths of controlled observation
- Replicable = easy to show reliability + generalisable after repeating.
- No confounding variables due to control
Limitation of controlled observation
- Reducing naturalness of environment + behaviour = due to regulating variables.
- Demand characteristics + low ecological validity = Participants know they’re being studied
Covert observation
Participants behaviour is watched and recorded WITHOUT their knowledge/consent.
Strengths of covert observation
- High ecological validity
- Removes participant reactivity = behaviour is natural as they’re unaware of the observation.
- Authentic
limitation of covert observation
- Unethical = lack of informed consent + deception
- Intrusive = lack of privacy
- No control over variables
Overt obervation
Participants behaviour is watched and recorded WITH their knowledge/consent.
Strength of overt observation
- Ethically acceptable = informed consent + the right to withdraw is given.
limitation of overt observation
- Hawthorne effect = altered behaviour due to awareness of observation.
- Demand characteristics affect data
Participant observation
Researcher becomes a member of the group being watched and recorded.
Limitation of participant observation
- Subjective and biased = observation made by someone that actively participated in the activity being observed.
- Researcher ‘goes native’= line between researcher and participant is blurred
Non-participant observation
Researcher doesn’t become involved with the group being watched and recorded.
eval of Non-participant observation
- Lack of direct involvement = objective and less likely to ‘go native’
- Easier to observe and record data.
limitation of non participant observation
Loss of valuable insight = may miss some things
Self reporting techniques
A method where a person is asked to state their own feelings, opinions and experiences related to a topic:
Example: questionnaires, interviews
Questionaires
A set of pre-written questions used to collect data by assessing a person’s thought or experiences.
- May be used to assess the DV
- Always structured
Advantages of questionnaires
Directly observing intentions/feelings = Reduces assumption.
- Can be distributed to a large sample = good for generalisation.
- May be more willing to share personal information.
Disadvantages of questionnaires
- People may be untruthful = due to social desirability bias.
- Completed by people that can read/write = limits sample.
- ‘Eager sample’ = filled by people who have time/want to fill them so the sample is biased.
- Acquiscence bias = tendency to agree to content without reading q’s properly.
State 3 guiding principles of questionnaires
The validity, objectivity and scientific nature is dependant on the design of the questionnaire:
- Clarity
- Bias
- Analysis
Describe the effects of Clarity in Questionnaires
Questions must be clear, high in face validity, shouldn’t be ambiguous and no use of double negatives:
- Double-barrel questions reduce clarity (e.g how satisfied are you with school AND your grades)
Describe the effects of Bias in Questionnaires
Any bias in a question may produce a biased answer:
- e.g how effective was the obnoxious president trump
Describe the effects of Analysis in Questionnaires
Questions must be simple enough to analyse.
Strengths and limitations of Open questions
- Allows people to elaborate on answers and be more detailed.
- Qualitative data BUT difficult to analyse
Strengths and limitations of Closed questions
Easy to analyse
- Lacks detail + forces you to choose an option (yes/no) = lacks validity.
State 3 ways to improve questionnaires
- Filler questions = have random unrelated questions embedded to reduce demand characteristics.
- Sequence of questions = easiest to hardest questions reduces resistance and anxiety in the person.
- Sampling techniques = alternative sampling method to overcome volunteer/eager sample.
Interviews
A method where data is collected face to face by an interviewer:
- Asks questions to assess interviewee’s thoughts/experiences.
Structured interview
Made up of a pre-determined set of questions that are asked in a fixed order.
Strengths of Structured interviews
- Easily repeatable = questions are standardised so different answers can be compared.
Limitation of structured interviews
- Validity of interview may be compromised = investigator effects due to their behaviour towards different interviewees.
- Less detail obtained.
Unstructured interview
Questions are developed during the course of the interview:
- Aims that a certain topic will be discussed + free flowing.
- Interviewee is encouraged to expand + elaborate on answer.
Strengths of Unstructured interviews
Higher validity due to more detail = due to follow-up questions
Limitations of Unstructured interviews
- Requires trained/skilled interviewers due to coming up with questions on the spot = expensive
- No pre-determined questions = questions may lack objectivity.
- Researcher has to sift through irrelevant info and draw conclusions = difficult and time-consuming.
- Social desirability = untruthful answers
Semi structured interview
Begins with pre-determined questions but follow-up questions are asked when appropriate.
State 3 guiding principles of Interviews
- Recording the interview
- Effect of the interviewer
- Interviewer bias
Describe the effects of Recording in Interviews
- Record audio + video and have the interview scribed.
- If the writer is required to write answers down, their listening may interfere with writing + vice versa.
- Writing/non-writing = may come across as intimidating or that the interviewee made a mistake.
Describe the effects of the interviewer in Interviews
- Presence of interested interviewer = yields fruitful + detailed answers = interviewers must be mindful on their non-verbal communication. (body language/posture)
- Mindful of verbal communication = knowing when to speak and not to speak.
Describe the effects of Interviewer bias in Interviews
When the interviewers expectations are unconsciously communicated and has an effect on the interviewees behaviour.
State 2 ways to improve interviews
- Avoid repeating questions
- Avoid probing and asking suggestive questions
= e.g. ‘are you sure?’
Describe 3 types of closed questions for Questionnaires
Likert scale = when the respondent indicates their agreement on a scale of strongly agree to strongly disagree.
Rating scales = when the respondent identifies a value that represents the strength of their feelings towards the subject from 1 to 5.
Fixed choice option = when the respondent is given a list of options and they tick what applies to them.
Describe 2 factors of designing an interview
Interview schedule = a list of questions that the interviewer intends of covering, this should be standardised between each participants to avoid investigator effect.
One to one interviews = increases the likelihood of the interviewee opening up, should start with neutral qs to get them relaxed and constantly reminded that everything they share is confidential.
State 3 methods for writing BAD questions
- Overuse of jargon
- Emotive language and leading questions
- Double-barrelled questions and double negatives
-ambiguous
Explain the effect of the Overuse of jargon
Overusing technical terms that are only familiar to people specialised in that field = confuses the respondent as it’s unnecessarily complex.
Explain the effect of Emotive language and leading question
- The author’s attitude to a topic may become clear through the way the question is phrased = ‘barbaric, destroyed, crashed’
- Leading questions guide the respondent to a particular answer.
Explain the effect of the Double-barrelled question and double negatives
- Contains 2 questions in 1
- The issue is that respondents may agree with half the question and not the other.
- Double negatives may be difficult to interpret
(I am not unhappy with my job agree/disagree)
Definition and evaluation of Qualitative data
Data expressed in words + non-numerical:
- More detail than quantitative
-Greater external validity = provides more insight to participants view.
- Difficult to analyse + subjective interpretations due to bias and the researcher may have preconceptions about what they’ll find
Definition and evaluation of Quantitative data
Data expressed numerically:
- Simple to analyse = comparisons between data are easy to find.
- More objective + less open to bias
- May fail to represent real life
Statistical test
Definition and evaluation of Primary data
Info that has been obtained 1st hand by the researcher:
- Authentic data that can be designed to specifically target info the researcher requires (e.g interviews)
- Requires more time and effort
Definition and evaluation of Secondary data
Info that has already been collected by someone else:
- Inexpensive + easily accessed with minimal effort
- May be variation in quality, outdated or incomplete
Describe how to work out the Mean and state a strength
Add all the values up, then divide by how many values there are.
- Includes all the values in the calculation = representative of all data.
Describe how to work out the Mode
The most frequently occurring value in a set of data
Describe how to work out the Range
Subtract the lowest value from the highest value
Definition of Standard deviation
A measure of how much scores deviate from the mean score:
- Low SD = data is tightly clustered around the mean, so all participants responded in similar ways.
- Large SD = not all participants were affected by the IV in the same way as data is widely spread + some anomalous.
interpretative validity
this is the extent to which the researcher’s interpretation of events matches
those of their participants.
Single blind procedure
participants are not made aware of the aims of the study until they have taken part
unaware of which experimental condition they. are in
(to reduce the effect of demand characteristics on their behaviour)
Double blind procedure and strengths
a third party conducts the investigation without knowing its main
purpose either
(which reduces both demand characteristics and investigator effects and
thus improves validity)
Internal validity
- this is whether the outcomes
observed in an experiment are due to the
manipulation of the IV and not any other factor
Ecological validity
This is the extent to
which findings can be generalised to other
situations and settings.
Temporal validity
Generalisability to other
historical times and eras
Population validity
Generalisability to different populations of various ages
Face validity
- this is when a measure is
examined to determine whether it
appears to measure what it is supposed
to. This can be done either through simply
looking at it or passing it to an expert to
check.
Concurrent validity
this refers to the
extent to which a psychological measure
compares to a similar existing measure.
The results obtained should either match or
be closely similar to the results of the well
established and recognised test.
Predictive validity
this refers to how well
a test can predict future events or
behaviours
E.g. how childhood attachment measured
using the strange situation are able to predict
how the child will grow up to behave in
adulthood (from Attachment topic).
Validity
refers to the extent to which results of a research study are legitimate. There are various
types of validity and ways of assessing them:
Meta-analysis
this is when a
researcher combines results
from many different studies
and uses all the data to form
an overall view of the subject
they are investigating.
strength of meta analysis
More generalisability is
possible as a larger amount of
data is studied.
- The researcher is able to
view the evidence with more
confidence as there is a lot of
it.
weakness of meta analysis
Publication bias such as the
file drawer problem may be
presented- this is when the
researcher intentionally does
not publish all the data from
the relevant studies but
instead chooses to leave out
the negative results. This
gives a false representation of
what the researcher was
investigating
Summarising data in a table
data has been converted into descriptive statistics
Bar Charts
- discrete data ( whole values e.g 3 cats not 3.5 cats)
-data that has been divided into categories
-do not touch each other which shows that we are dealing with separate conditions
freq : y-axis
categories : x-axis
Histograms
-continous data (doesnt have to be full numbers e.g 3.75ml of water)
freq: y-axis
catgories : x-axis
Line graphs
- continious data
y-axis : DV
x-axis : IV
scattergrams
- associations betweeen co-variables
-rather than differences hence we
came across them in the correlations topic
-Either of the co-variables can occupy the x-axis or
the y-axis,
normal distributions
a symmetrical pattern of frequency data that forms a bell-shaped pattern
skewed distribution
a spread of frequency data that is not symmetrical, instead the data
all clusters to one end.
positive : distr of data is concentrated on the right
negative : distr of data is concentrated on left
Define Peer review
the assessment of scientific work by experts in the same field
- it is done to make sure that all research intended to eventually be published is of high quality.
What are the main purposes of peer review?
- to know which research is worthwhile so that funding can be allocated towards it
- to validate the relevance and quality of research : prevent fraudulent research from being publicised
-to suggest possible improvements for the study
what are real life examples of why papers are peer reviewed?
- peer review contributes to research rating of university departments
-public journals must be peer reviewed before publication
Limitations of peer review
Publication bias: editors tend to prefer to publish positive results or “headline grabbing” results as opposed to any negative results :
file drawer problem , where negative results are intentionally not published
- negative findings are important so replications can be made to check validity
Once a study has been it is often difficult to retract , even when proven to be wrong : once in public domain the damage is done
It can be difficult to find an expert.
Smith (1999) argues that because of this a lot of poor
research is passed as the reviewer didn’t really understand the work.
Outline real life consequences of peer review
MMR VACCINE : (Andrew wakefield 1998) the vaccine leads to autism: had implications as number of measles cases increased : later found research was fraudulent
- however rumours about MMR continue to persist
11+ EXAM
How is psychology and the economy linked?
how what we learn from psychological research
influences our country’s economic prosperity
mental healthy: £22 spent on MH annually in the UK e.g stress anxiety
For such problems psychology research has been able to present solutions to them and this expresses why psychology research is important for the economy.
economical implications : psychopathology
Treatments - Cognitive Behavioural Therapy and Rational Emotive Behavioural Therapy for depression, drug therapy for OCD.
Economy - workers able to return to work
economical implications : Attachment
Role of the father - Tiffany
Field (1978) found that fathers
can take on the role of being a
primary caregiver.
Economy - - Mothers can return to work.
- More flexible working arrangements within families.
- Can maximise their income and effectively contribute to the economy.
economical implications : social influence
Social influence leading to social change -Minority influence, appealing to NSI, disobedient models.
Economy :
-Health campaigns.
- Unions strike- make working conditions better.
- Environmental campaigns like getting companies to reduce their waste and use of non-renewable energy.
economical implications : memory
Eyewitness testimony - How leading questions or post event discussion can affect
eyewitness testimony.
Economy - Led to police using the
cognitive interview which reduces wrongful convictions hence reduces waste of money and space in jail.
Case studies
detailed study into the life of a person which covers great detail into their background
- builds up a case history hence providing qualitative data
Examples an example of case studies
case of HM : memory:could tie shoelaces but couldn’t remember stroking a dog : procedural memory intact but episodic wasn’t
Strengths of Case studies
- Detailed so able to gain in depth insight
- Forms basis for future research.
- From studying unusual cases you are able to
infer things about normal usual behaviour of
humans. - Permits investigation of situations that would
be otherwise unethical or impractical.
Limitation of case studies
- Not generalisable to wider populations as
data is only gathered from one person. - Various interviewer biases are presented like
social desirability bias (from the unique
person’s side) and interpretative bias ( from
the researcher’s side). - They are time consuming and difficult to
replicate.
content analysis
qualitative research tool or technique widely used to analyze content and its features.
- This allows us to have insight into the
structured values, beliefs and prejudices of our society.
how to conduct a content analysis
● Identify hypothesis that you will investigate.
● Create a coding system depending on what you are investigating e.g. 1= male, 2= female.
● Gather resources.
● Conduct content analysis and record data in a table.
● Analyse data which is descriptive and qualitative e.g. using ‘thematic analysis’- allows
themes, patterns and trends to emerge in data.
● Write up a report in the format of a scientific report.
Strengths of content analysis
- Strong external validity as the data is already
in the real world so it has high mundane
realism. - Produces large data set of both quantitative
and qualitative data that is easy to analyse. - Easy replication.
- Ethical issues like ‘right of privacy,
confidentiality, informed consent’ are avoided
as data is already in the public domain.
Limitations of content analysis
- Observer bias is presented but it can be
eliminated by achieving inter-observer
reliability. - Content of choice to analyse can be biased
by researcher. - Interpretative bias - the researcher may
ignore some things but pay extra attention to
others.
what are the 3 levels of measurement of quantitative data
- nominal
- ordinal
- interval
- ratio
Nominal data
- type of data that is in the form of categories.
- Qualitative data
- discrete- one item can only appear in one category.
- difference between categories have no further meaning
It does not enable sensitive analysis as it does not yield a numerical result for each participant
Ordinal data
- data represented in a ranking form e.g 1 = hates maths 10= loves maths
- no equal intervals between each unit
- Qualitative data
weakness : lacks precision : based on subjective opinions of people
interval data
-named and ordered
based on numerical scales which include equal units of precisely defined size
- Quantitive data
-Continuous data
Ratio
- same characteristic as interval data
- however has a meaningful ZERO point
e.g time taken to undertake a task
Outline appropriate measures for each level of data
Nominal: Mode, no measure of disperesion
Ordinal: Median , range
Interval: Mean , SD
Scientific report
- writing up of a research for publication
- has a specific series of sections
Outline the sections of a scientific report
Title : what is the report about?
Abstract : brief summary of study
Introduction : background of study
Method : process of study
Results : summarise findings
Discussion : findings and their implications
**References ** : inform reader about sources of information used
Appendices: detailed info not in report
Abstract
- concise summary of report
-key details of report
-150-200 words long - includes : aim , hypothesis , method , results and conclusion
- read to know whether research study is worth examining any further
Introduction
information of past research on a similar topic whereby relevant theories , studies and concepts are mentioned
broad leading towards specific detail
allows reader to place the study in context
Method and what is included
Description of how the study was conducted
- Must be enough information to be able to replicate the study
includes : design, sample collected , materials used and procedure , ethics etc
Results
findings of study , presented with
- descriptive statistics : visual representation of differences between group
- inferential : analysis of data and null hypothesis accepted or rejected
Discussion
considering what the findings exactly mean and for psychological theories
-implications of the research such as on the society
- limitations and improvement
Refrencing
List all of sources that were quoted or referred to in the report
Appendicies
copy of all resources/ material used in the study , raw data and statistical calculations
(aids in peer review and replication)
Statistical testing
provides a way of determining whether hypotheses should be rejected or accepted
- tells us whether differences or relationships between variables that have been found during the xperiment are statistically significant or if they only occur due to chance
Alternative hypothesis (H1)
there IS a statistical difference
- there is a difference between the conditions
- small probability that the results are due to chance
Null hypothesis (H0)
there is NOT a statistical difference
-there is no difference between the conditions
- high probability that results are due to chance
When to use (Sign test)
- looking for difference not association
- used a related experimental design - repeated measures design
-collected nominal data
How to find value for sign test
- calculate difference between the two sets of data : minus them from eachother
-state if theyre a - or + - amount of negative data = y
- amount of positive data = x
x-y = sign value
N= no of participants , not including the participants whos difference is a 0
-FIND CR on table
how to conduct sign test
1) hypothesis : alternative and null hypothesis
2) record data and work out the sign
e.g sign will be negative (-) if the
value has decreased in the second condition but positive (+) if it has increased. If the value
has stayed the same , this value will be ignored and the N adjusted to exclude it.
3) calculated value for sign test , S,
4) find critical value of S - use calculated N value and p<0.05 which means that theres less than 5% prob thhat results occured by chance
5) conclusion , refer back to hypothesis mentoning IV and DV in conclusion etc
what are the 3 factors you must take into consideration when choosing an inferential statistical test?
Design of study:
- unrelated design ? - independant group design
- related design? - repeated measures or matched pair design
level of data collected during study: ordinal , nominal or interval
difference or correlation being measured
when to use chi squared test
Nominal + unrelated : chi squared
Nominal + related : sign test
test for association/correlation
What data’s should i use
Nominal + unrelated : chi squared
Nominal + related : sign test
test for association/difference : chi-squared
Ordinal + unrelated : Mann-whitney test
Ordinal + related : Wilcoxon test
test for association/difference : spearman’s Rho
Interval+unrelated : unrelated t-test
interval+ related= t-test
test for association/difference : Pearsons R
critical value
the numerical boundary that stands between accepting or rejecting the null
hypothesis when a hypothesis is being tested
event sampling
Event sampling is used to sample behaviour in observational research.
Rule of R
If there is an R in the name of the statistical test the calculated value has to be
gReater or equal to the critical value for the result to be significant.
what are the 2 types of errors that can occur during inferential statistical testing
Type I : incorrect rejection of null hypothesis (false positive)
- therefore findings were actually due to chance
Type II : incorrect accepting of null hypothesis and rejection of alternative
(false negative)
-therefore findings were actually significant
paradigm and example
set of shared ideas and assumptions within a scientific discipline about a subject and its method of enquiry
e.g Kuhn argues that a subject can only be called a science if there is an agreed global theory
e.g evolution
paradigm shift
a significant change in these central assumptions within a scientific
discipline, resulting from a scientific revolution.
outline 3 stages of a science
pre science : no paradigm exists , much debate about what the subject is and its theoretical approach
normal science : generally accepted paradigm can explain and interpret all findings
scientific revoluution : evidence against the old paradigm reaches a certain point and there is a paradigm shift . Old paradigm is replaced by new one
What scientific stage is psychology at?
- pre science
- too much disagreement and conflicting approaches
Theory
set of general principles and laws which can be used to explain specific
events or behaviours
Theory construction
gathering evidence from
direct observation during investigations.
- order and direction in science
Deduction
deriving new hypotheses
from an already existing theory
e,g Baddeley and Hitch modifying WMM in 2000 as they added the episodic buffer to model
Falsifiability
-principle that states a theory cannot be considered scientific unless it
allows itself to be proven untrue
- (Popper)states that successful theories that
have been constantly tested and supported simply haven’t been proven false yet
-‘pseudosciences’ : sciences that cant be proven wrong e.g Freud
the alternative hypothesis is always
accompanied by the null hypothesis.
-formulating hypotheses that can either be proved or disproved by experimentation.
Replicability
-the extent to which scientific methods and their results can be repeated by other researchers across other contexts and circumstances.
- validity and reliiability of results
Objectivity
all possible biases from the researcher are minimised so that they don’t influence or distort the research process.
empirical methods of investigation :
making direct observations and through direct experiences.
- cant be scientific if not empirically tested and verified using either empirical or observational methods
Strengths of psychology as a science
- Scientific methods are used in many research
studies giving them scientific credibility. - Findings from studies do positively impact
society & individuals e.g. Cognitive behavioural
Therapy to treat depression.
Limitations of psychology as a science
- Not all research is generalisable e.g. from
case studies. - Psychologists do often make inferences of
behaviour rather than directly measuring it , for
example this is usual for cognitive
psychologists that infer about cognitive
processes from brain scans (Memory topic link
here).
Definition of ethics in psychology
A matter of balance between the rights of participants and the goal of research to produce valid data.
Explain the BPS code of ethics
Instructs psychologists in the UK about acceptable and unacceptable behaviour when dealing with participants.
- Based on 4 major principles: respect, competence, responsibility and integrity
State 6 ethical issues in research: Can Do, Can’t Do With Participants
- Informed Consent
- Deception
- Confidentiality
- Debriefing
- Right to withdraw
- Protection from physical and psychological harm
Explain ways to deal with Consent
Participants must be fully informed of the aims of the experiment, procedures, risks and their rights
-They’ll choose whether to participate without being coerced.
- Consent forms should be used: under 16s and medically ill must be consented by guardians.
Explain the use of Deception in research
Deliberately misleading/withholding information = invalid consent, but may be necessary in some cases
Describe the use of Debriefing after Deception
After the study, participants should be fully informed of the true aims of the experiment and must be able to Withdraw their data.
- Should be allowed to express concerns/questions + be provided with counselling if subjected to stress.
Describe the use of the Right to withdraw in research
Participants must be allowed to withdraw themselves and their data at any time during the study.
Explain ways to deal with Confidentiality
Identities of the participants should remain private and unidentifiable in published research.
- Participants must be able to control information about themselves = right to privacy
Explain ways to deal with Protection from physical and psychological harm
Participants shouldn’t be placed at any more risk than they would be in their daily lives.
- Includes: embarrassment, stress and pressure
- They should be constantly reminded of their right to withdraw, counselling and therapy in extreme situations.
Measures of central tendancy
Mean
Mode
Median
Measures of dispersion
Range
SD
Percentages
posittive neegative and zero calculattions
Pilot study
small-scale version of an investigation which is done before the real
investigation is undertaken
-so that participants do not bias results in ways that they “think” they should act e.g demand characteristics
Timed sampling
a method of sampling behaviour in an observation study and is where an observer records behaviour at prescribed intervals
timed sampling evaluation
+ Less likely to miss behaviours as the researcher usually has a short time to focus on recording behaviour, therefore is more likely to be accurate.
-Behaviours that occur outside the time intervals are not accounted for, therefore may reduce validity as important behaviours may be missed.
how to increase reliability of observations
Check inter-rater reliability
Conduct a pilot study to check behaviour categories
controlled observational study
Watching and recording behaviour within a structured environment
i.e. one where some
variables are managed
Controlled observational study eval
+ Replication is easy as less extraneous variables
-Cannot be applied to real life settings and low in ecological validity
Naturalistic observation
Observation of behaviour in a natural setting
- Investigator does not interfere, just observes + records
Naturalistic observation evaluation
+ High in ecological validity
+ Used to generate ideas for experimental research / validate experimental findings
-No manipulation of variables = cannot infer cause+effect
-Lack of control = no replication
-Ethical problems = invasion of privac
Behavioural categories
when psychologists must decide which specific behaviours should be examined
these should be observable/objectively defined/operationalised/unambiguous.
“behaviour checklist” affection : hugging kissing etc
Efficient behavioural categories
must be observable efficient and self evident
Event sampling
involves counting how the number of times a particular behaviour occurs within a target individual or group
Sampling methods eval ( event and timed sampling)
+less time consuming : reducing the number of observations that have to be made
-if specified event is too complex the observer may overlook important details using event sampling
test retest reliability
a way of assessing the external reliability of a research tool. It involves presenting the same participants with the same test or questionnaire on two separate occasions, and seeing whether there is a positive correlation between the two.
