Research methods Flashcards
define aim
a statement which explains what the experiment is attempting to achieve
define hypothesis
a clear, precise, testable statement which makes a prediction of the direction that results may take
what is the difference between an aim and hypothesis
an aim is a statement of investigation whereas the hypothesis is a prediction
what are the two types of hypothesis
alternative- predicts there will be a difference/relationship
null- predicts there will be no difference/relationship
what are the two types of alternative hypothesis
directional- predicts the direction the results will go
non-directional- predicts a difference but not a direction
define the independent variable
the variable being manipulated
define the dependent variable
the variable being measured
define extraneous variables
additional or unwanted variables which are, where possible, controlled or removed by the researcher
define confounding variables
variables which change the iv
define demand characteristics
participants behave in the way they believe is expected of them or find the aim and behave in a way to influence the results
define the hawthorne effect
being watched makes us change how we act
define the social desirability bias
acting in a way to look ‘good’
define investigator effects
a researcher consciously/unconsciously influencing the outcomes of the research
types of investigator effects
non verbal communication- body language hints at correct/incorrect behaviour
physical characteristics- behaviour changes based on the appearance of the researcher
bias in interpretation of data- interprets what is wanted to be obtained from the data
define order effects
the order that participants complete conditions in affects participants’ behaviour
what are the two types of order effects
practice effects- participants may perform better in the second condition as they know what to expect
fatigue effects- performance may be worse in the second condition as participants are tired
how can order effects increase the chances of demand characteristics
practice effects- participants find the aim with repeated conditions, performing better and affecting results
fatigue effects- participants may become bored ad so affect the study by manipulating results
describe standardisation
keeping variables the same
describe counterbalancing
ensuring variables occur in all possible combinations
describe randomisation
deciding the order of the variables by chance
describe single blind
ensuring that participants do not know the details of the experiment
describe double blind
participants and experimenters do not know the details of the experiment
describe independent groups
ppts only complete one condition of the experiment
strengths of independent groups
+no order effects
+lower drop-out rate (attrition)
weaknesses of independent groups
-groups may not be well matched
-requires twice as many ppts
describe repeated measures
all ppts complete all conditions of the experiment
strengths of repeated measures
+less ppts saves money and time
+groups are perfectly matched
weaknesses of repeated measures
-may show order effects
-demand characteristics may affect results
describe matched pairs
ppts are matched based on a factor and each complete a condition
strengths of matched pairs
+control of participant variables
weaknesses of matched pairs
-time to gather and match participants
-may not be possible to completely match both groups
define target population
the total group of individuals from which the sample might be drawn
define sample
a group of people who take part in research who are presumed to be fair and representative of the total population
define sample bias
certain groups are over/under represented within the sample selected, limiting generalisability to outside the sample
define representative sample
a sample that will reflect how the population as a whole would respond to the experiment
describe random sampling
all of the target population’s members have an equal chance of being selected
strengths of random sampling
+no researcher bias
+prevents choosing ppts to support hypothesis
weakness of random sampling
-time consuming
-still unrepresentative
-participants may refuse to take part
describe systematic sampling
there is a system for selecting participants, every nth person
strengths of systematic sampling
+no researcher bias
+representative
weaknesses of systematic sampling
-time consuming
-unrepresentative if ppts are underrepresented
describe stratified sampling
ppts selected based on their proportion in the population
strengths of stratified sampling
+avoids researcher bias
+representative
weakness of stratified sampling
-complete representation is not possible
-time consuming
describe opportunity sampling
ppts selected based on who is present when conducting the study
strengths of opportunity sampling
+saves time and effort
+less costly
weaknesses of opportunity sampling
-researcher bias
-unrepresentative of the target population
describe volunteer sampling
ppts are collected by asking people to take part in the study
strengths of volunteer sampling
+less likely to drop out
+easy and quick
weaknesses of volunteer sampling
-volunteer bias
describe a lab experiment
performed in a controlled environment, using standardised procedures with ppts randomly allocated to groups
strengths of lab experiments
+good control = easy to replicate
+cause and effect can be established
+gathers quantitative data
weaknesses of lab experiments
-artificial setting = low ecological validity
-ppts more likely to display demand characteristics
describe a field experiment
occur in real world settings, the IV is manipulated by the researcher and as many variables as possible are controlled
strengths of field experiments
+natural setting = ppt behaviour is more likely to be natural
+less chance of demand characteristics
weaknesses of field experiments
-ethical issues, ppts may not know they are in an experiment
-cannot control all variables, EVs may affect results
describe a natural experiment
the IV varies naturally, the experimenter does not manipulate it but measures the effect on the DV
strengths of natural experiments
+allows researchers to investigate a behaviour not normally allowed to be studied
+less chance of demand characteristics, ppts unaware they are in a study
weaknesses of natural experiments
-must wait for the IV to occur
-less control as the IV cannot be manipulated
describe a quasi experiment
the IV occurs naturally, the experimenter is unable to change the IV or randomly allocate ppts to conditions
strengths of quasi experiment
+allows researchers to investigate behaviour that wouldn’t normally be allowed to be studied
+can be some control of EVs
weaknesses of quasi experiments
-must wait for the IV to occur
-less control as the IV cannot be manipulated
describe naturalistic observations
the setting has not been manipulated, the set up is natural
strengths of naturalistic observations
+higher ecological validity
+natural, spontaneous behaviour
limitations of naturalistic observations
-less controlled, EVs could affect the results
describe controlled observations
some variables have been manipulated
strengths of controlled observations
+more controlled, can be certain that behaviour is due to the situation
limitations of controlled observations
-lower ecological validity
- ppts less likely to act in a natural way
describe covert observations
observations are made without the ppts knowledge
strengths of covert observations
+ppts will behave more naturally
limitations of covert observations
-ethical issues
-observers may misinterpret behaviour
describe overt observations
ppts are aware that they are being observed
strengths of overt observations
-avoids ethical issues
limitations of overt observations
-observer effect, ppts may alter behaviour as they know they are being watched
describe participant observations
observer is a ppt in the behaviour being observed
strengths of participant observations
-likely to better understand behaviour
limitations of participant observations
-observer bias, less objective, harder to record seen behaviour
describe non-participant observation
observer is not a ppt in the behaviour being observed
strengths of non-participant observations
+more objective, distanced from behaviour
describe structured observations
there are behavioural categories and/or sampling methods
strengths of structured observations
+improved inter observer reliability
limitations of structured observations
-observer bias, may see what they want to see
describe unstructured observations
there are no behavioural categories or sampling methods
strengths of unstructured observations
+useful when behaviour is unpredictable
+good as a pilot study to make categories
limitations of unstructured observations
-eye catching behaviours may not be the most relevant
-poor inter observer reliability, no established categories
define behavioural categories
a list of specific behaviours that are counted every time the behaviour occurs
describe time sampling
behaviour is recorded at fixed time intervals
strengths of time sampling
+shows time related changes in behaviour
weaknesses of time sampling
-behaviours can be missed when they happen between time intervals
describe event sampling
behaviour is recorded every time it is seen
strengths of event sampling
+useful when recording behaviours that don’t happen frequently
weaknesses of event sampling
-behaviours may be missed if they are happening too frequently to count
define inter observer reliability
a measure used to assess the degree to which different observers give consistent estimates to the same phenomenon
how do you know if inter observer reliability is high
good inter observer reliability has a correlation of 0.8 or higher (strong positive)
define a questionnaire
a set of written questions used to assess a person’s thoughts/feelings/opinions
strengths of questionnaires
+cheap and quick way to gather a large amount of data
+can be completely privately/anonymously so responses are more likely to be honest
weaknesses of questionnaires
-social desirability
-flawed if there are leading questions
-questions can be misunderstood
what are the three types of interview
structured, semi-structured, unstructured
describe structured interviews
a set list of questions to be answered, predetermined, in a fixed order, closed questions to generate quantitative data
describe unstructured interviews
no set questions, more conversational, open-ended questions to generate qualitative data
describe semi structured interviews
usually first question is pre-set, follow up questions are decided based on what has been said, most common type
strengths of structured interviews
+easy to collect data
+easy to replicate and compare
weaknesses of structured interviews
-does not allow for elaboration on answers, less qualitative data
strengths of semi/unstructured interviews
+detail in specific areas
weaknesses of semi/unstructured interviews
-difficult to replicate
describe an interview
formal conversations between an interviewer and respondent where the interviewer leads a question-answer session
define open questions
allow respondents to provide as much or as little as they wish in response to a question
strengths of open questions
+insight into thoughts/feelings/opinions
+qualitative data
weaknesses of open questions
-difficult to analyse
define closed questions
respondents are given a set of predetermined responses
strengths of closed questions
+can be analysed and compared
+quantitative data
weaknesses of closed questions
-does not give insight into or reasons for responses
what are the characteristics of good questionnaire questions
not leading (non-biased/judgmental), balance of open/closed
describe focus group interviews
a qualitative approach where a group is interviewed together to gain an in-depth understanding
strengths of focus group interviews
+more people interviewed in a shorter time
weaknesses of focus group interviews
-some people may speak more than others
-some may conform to group ideas (social desirability bias)
describe a correlation
correlations look at the relationship between variables
define correlation coefficients
a figure calculated from a statistical test that determines the strength and direction of a relationship, on a scale from -1 to +1
strengths of correlations
+good indication of how variables are related
+starting point to assess links between variables before an experiment
+economical, good use of secondary data
+predictions can be made
+quantification of relationships
weaknesses of correlations
-no cause and effect
-possibility of untested/intervening variables causing the relationship
-often misinterpreted
define case study
a longitudinal in-depth study of a person, group or event, usually of abnormal occurrence
strengths of case studies
+provides rich, in-depth data
+suggests directions for further research
+allows investigations that might be difficult or unethical to do in other ways
weaknesses of case studies
-problems generalising results to the wider population
-loss of detachment and objectivity
-difficult to replicate and time-consuming
define informed consent
ppts know what to expect from the experiment and have agreed to take part
dealing with informed consent
researchers must outline the study and ask ppts for consent. if unable to do this, researchers can get presumptive consent (how ppts would feel)
define deception
researchers misleading or lying to ppts about a key aspect of the study
dealing with deception
should be avoided, or ppts deceived as little as possible. should be revealed as soon as possible after
define confidentiality
keeping ppts data anonymous and protected
dealing with confidentiality
info about ppts should be kept in a locked, safe space and details kept anonymous
define debrief
informing ppts about the experiment afterwards, answering questions and providing support if needed
dealing with debrief
should take place as soon as possible after the research, explaining info and providing support resources
define right to withdraw
ppts are allowed to remove themselves and/or their data at any point during research
dealing with right to withdraw
ppts should not have pressure placed upon them to continue, should be aware that withdrawing is possible at any stage
define protection from harm
ppts should not be subjected to any physical or mental distress
dealing with protection from harm
harm must be no greater than in everyday life, vulnerable groups must receive special care. therapeutic debriefing can be used if distress has been caused
define quantitative data
numerical data
define qualitative data
descriptive data, usually about thoughts, opinions, feelings
strengths of quantitative data
+easy to analyse and compare
weaknesses of quantitative data
-not descriptive
strengths of qualitative data
+descriptive, insight into behaviours
weaknesses of qualitative data
-difficult to analyse
define primary data
data collected first hand for the purpose of the study
strengths of primary data
+more tailored to research
weaknesses of primary data
-time consuming
-expensive
define secondary data
data collected second hand, for another purpose/in another study
strengths of secondary data
+cheaper
+time saving
weaknesses of secondary data
-less tailored to research
define meta analysis
collation of multiple pieces of research
strengths of meta analysis
+increases validity, using a wide range of research
weaknesses of meta analysis
-less suitable to research
strengths of the mean
+valid measure
limitations of the mean
-can be skewed by outliers
-decimal scores are meaningless
strengths of the median
+cannot be skewed
limitations of the median
-less representative, does not take in all numbers
strengths of the mode
+cannot be skewed
limitations of the mode
-crude measure
-isn’t always a mode
strengths of the range
+easy to calculate
limitations of the range
-not a very useful method
-doesn’t take much data into account
define standard deviation
uses all the scores in data to calculate the spread of scores and how much each score deviates from the mean
strengths of standard deviation
+all data taken into account
+precise measure of dispersion
limitations of standard deviation
-can be affected by extreme values
define peer review
the assessment of scientific work by others who are specialists in the same field to ensure that any research intended for publication is of high quality
what are the three aims of peer review
- to allocate research funding
- to validate the quality and relevance of research
- to suggest amendments or improvements
describe the method of peer review
- study is conducted, and a report created which is submitted for peer review
- the paper is sent to psychologists specialised in the same field as the paper who judge it on its credibility, validity, and appropriateness
- peers give feedback and make a recommendation of publication
strengths of peer review
+maintains high standards in research
+contributes new knowledge to the field
limitations of peer review
-if anonymity is not maintained, bias may affect feedback
-file drawer effect, only statistically significant findings are published
describe the process of reporting psychological investigations
title, constants, abstract, introduction, aims, hypotheses, method, results, discussion, conclusion, references, appendices
describe an abstract
summary of the study, outlines the aim, method, results, conclusion, written at the end of the report
stats test table
nominal, related- sign test
ordinal, related- wilcoxon signed rank
interval, related- related t
nominal, unrelated- chi squared
ordinal, unrelated- mann-whitney
interval, unrelated- unrelated t
nominal, correlation- chi squared
ordinal, correlation- spearman rho
interval, correlation- pearson product
what are the 3 types of distribution
normal, positive, negative
describe normal distribution
the mean, median and mode are all similar values
describe positive distribution
the mean, median and mode are pulled to the left with more of a tail to the right
describe negative distribution
the mean, median and mode are pulled to the right with more of a tail to the left
what reasons are there for skewed data?
- difficulty of test
- anomalous results
- sample/question bias
what does p<0.05 mean (in terms of level of significance)?
there is a 5% or less chance that a factor other than the IV has caused the results, we can be 95% sure that the IV directly affects the DV
why do we choose a 5% level of significance?
good compromise- not too lenient (not left to chance) or too strict (findings can be accepted as significant)
explain type 1 errors
margin is too lenient, results show a false positive, alt hypothesis has been accepted when it should have been null
explain type 2 errors
margin is too strict, results show a false negative, null hypothesis accepted when it should have been alt
what are the three types of inferential statistics data?
nominal, ordinal and interval
describe nominal data
named categories, not much detail, MODE
describe ordinal data
rank ordered, vague, MEDIAN
describe interval data
scientific measurement, each unit is the same size, MEAN
how to do a sign test?
- work out if positive or negative difference between 2 data sets
- exclude any 0 difference
- count the number of positive and negative signs
- take the smallest value as the calculated value
what do you need to find the critical value?
-direction of hypothesis
-number of ppts (degrees of freedom)
-level of significance
what may affect number of participants/data sources when identifying critical values?
-any ppts excluded for having 0 difference
-degrees of freedom in chi squared, related/unrelated and pearson’s test
degrees of freedom formula- chi squared
df = (rows in table - 1) x (columns in table - 1)
what correlation co-efficient is a reliable result?
0.8+
define inter observer reliability
two researchers using the same interview format, equipment or observation schedule should obtain the same results
describe how to check inter observer reliability
- decide behavioural categories together
- observe the same behaviour at the same time
- compare results of observation
- stats test to determine correlation
define test-retest
a method to test external reliability, same test is done by same people twice at different times, if results correlate then the test is valid
describe test-retest method
uses the same instrument on two different occasions with the same people
to check results, complete a correlation via a stats test (spearman’s or pearson’s)
define reliability
measuring variables consistently and getting the same results
define internal reliability
the consistency of a measure within itself
define external reliability
the consistency of a procedure from one occasion to the next
define validity
whether the items in an assessment are measuring what they are supposed to measure
define internal validity
whether the IV has had an effect on the DV, or if it is due to EVs
define external validity
the extent to which findings are generalisable beyond the research itself
what are the three types of external validity
ecological, population, temporal
define ecological validity
whether results obtained in an experiment can be extended to real life settings
define population validity
whether findings from a sample can be generalised to other types of people
define temporal validity
whether findings can be generalised to other time periods
define face validity
the extent to which a test is subjectively viewed as covering the concept it claims to measure (does it look valid)
define concurrent validity
a new measure is tested against a well-established one, correlation value between them determines validity
what are the 4 key features of a science
empirical methods, control, replicability, objectivity
define empirical methods
all knowledge should be gathered from direct, sensory experience
define control (features of a science)
controlling EVs in order to establish causation
define replicability
the ability to be consistent in measurement and results in order to draw conclusions from research
define objectivity
no personal biases which may impact data or behaviour of participants
what are the features of the scientific method
theory construction, hypothesis testing, falsifiability, paradigms
define hypothesis testing
the process of testing theories and hypotheses via deductive processes such as experiments
define theory construction
the process of creating and refining a theory based on inductive processes such as observation
define falsifiability
the ability to prove/disprove a theory/idea based on scientific data or experimentation
define paradigms
belief that a subject can only be called a science if the majority of its workers agree with a common global theory
what characterises the normal science phase of a paradigm
paradigm has been established, theory proposed that explains facts and unites field, researchers explore paradigm
what are the phases of developing a paradigm
pre science, normal science, scientific revolution
what characterises the scientific revolution phase of a paradigm
vast amounts of testing, evidence to contradict paradigm, new paradigm established
describe the steps of theory construction
- make observations
- produce laws/theories
- generate hypothesis
- test with experiments
- reject/refine theory
describe the inductive process
begins with a set of empirical observations, seeking patterns in observations and then theorising about patterns (steps 1-3 theory construction)
describe the deductive process
begins with a theory, developing hypothesis from theory, collecting and analysing data to test hypothesis (steps 4 and 5 of theory construction)
define descriptive statistics
data analysed to show patterns and trends, not significance
define inferential statistics
data analysed to work out statistical significance- if results are dependable or due to chance
