Research Methods Flashcards
What is the IV
The variable that is changed (manipulated)
What is the DV
The variable that is measured by the researcher
Define operationalisation
Clearly defining variables in terms of how they can be measured
Define extraneous variable
Any variable other than the IV that can affect the DV
What are confounding variables
Variables that are not the IV that have had an effect on the DV because they haven’t been controlled -thus confounding or ruining the results of a study making the results unreliable
What are situational variables
Things that are connected to the research situation which could affect the DV so should be controlled
What are participant variables
Things that are connected with the research participant which could affect the DV so should be controlled
Define aims
General statements of what the researcher intends to investigate the purpose of the study
Define hypothesis
A clear precise and testable statement that states the relationship between the variables to be investigated
What is a null hypothesis
One that always predicts that any difference or relationship is down to chance so there is NO significant difference
What is an alternative hypothesis
What are the two types
One that always predicts that something will happen
Non-directional and directional
Explain the difference between a directional and a non directional hypothesis
A non-directional hypothesis states that there is a difference but does not state the difference directly
Directional hypothesis clearly defines the difference that there will be between the two conditions (operationalises the outcome)
Briefly state the four different types of experiment
Lab experiment
Natural
Field
Quasi
Describe a lab experiment (evaluate)
The variables are highly controlled
Participants know they are taking part
Artificial
(Pro: establish cause and effect/can be replicated)
(Con: lacks ecological validity/ demand characteristics)
Describe a field experiment (evaluate)
Some control over variables
Participants do not know they are taking part
Real life
(Pro: high ecological validity/ less chance of demand characteristics)
(Con: cannot be replicated/ cannot establish cause and effect)
Describe a natural experiment (evaluate)
No control over variables Participants do not know they are taking part Real life (mostly) (Pro: high ecological validity/less chance of demand characteristics) (Con: cannot establish cause and effect/ cannot be replicated)
Describe a quasi experiment (evaluate)
Not strictly an experiment
Variables just exist
(Pro: can establish cause and effect/ replicated)
(Con: lacks ecological validity/ demand characteristics)
Briefly name and explain the three experimental group designs
Independent group design= different participants do each condition (one each)
Repeated measures= all participants do both conditions
Mixed pairs design= different participants are matched to one another and each placed in separate conditions
Evaluate independent Group design
✅Reduced demand characteristics
✅No order effects
❌Participant variables
❌More participants needed
Evaluate repeated measures design
✅No participant variables
✅Less participants needed
❌Demand characteristics
❌Order effects
EValuate matched pairs design
✅Reduced participant variables
✅No order effects
❌More participants needed
❌Difficult to match exactly
Name for ways that variables can be controlled
Counterbalancing
Standardisation
Randomisation
Random allocation
Explain what is meant by counterbalancing
It does not remove or prevent order effects but attempts to balance out the effects of order between two conditions
Half the participants take part in condition A then B the other take part in conditions B then A
used ONLY IN repeated measures
Explain what is meant by standardisation
All participants are subjected to the same environment information and experience
Same instructions and tasks
Explain what is meant by randomisation
To reduce bias when designing materials and deciding the order of conditions
minimise the effect of extraneous variables
Explain what is meant by random allocation
To address the problem of participant variables ONLY in an independent group design participants should be randomly allocated to different conditions
Define demand characteristics
Participants revealing the purpose of the investigation and changing their behaviour
Define generalisation
Extent to which findings and conclusions from an investigation can be applied to the population
What is volunteer sampling and evaluate
Participants select themselves to be part of an experiment
✅Less time-consuming
❌Volunteer bias
What is opportunity sampling
Researcher selects those who are most readily available
Pro= easy less time consuming Con= researcher bias
What is random sampling and evaluate
Researcher picks names out of hat of those in the target population RANDOMLY
Pro=unbiased
Con=does not guarantee a representative sample
What is systematic sampling
Gather names of those in target population and use a system, e.g every other 6 people are selected
Pro=avoid researcher bias
Con=not objective (researcher chooses how people are listed etc)
What is stratified sampling evaluate
Identify different sub groups in population, work out the ratio needed, randomly sampling participants from each sub group
Pro=very representative/ no researcher bias
Con= time consuming
When is a orrelation used
When there is a relationship
Evaluate using a correlation
✅Good preliminary tool for research=establish a link
✅used to study behaviour when it would be unethical to do so in a experiment
❌does not show cause and effect
Explain the difference of covert and overt
Covert is observing people without their knowledge
Overt is observing people who know they are being studied (pro=no ethical issues over privacy etc)
Explain the difference between participant or non participant
Evaluate this
Participant= observations are made by someone who is also a participant
Pro=good insight / con=less objective- too involved
Non participant= the observer is separate from the people being observed
Pro=more objective/ less insight (outsider)
What is the difference between event or time sampling
Event sampling involves counting the number of times a behaviour occurs
Time sampling= counting the behaviour within a fixed time frame
Define content analysis
Analysing qualitative data (turning into quantitative)
Evaluate content analysis
Pros:less ethical issues
Ecological validity
Replication
Cons: observer bias
Name three measures of central tendency
Mean
Median
Mode
What are descriptive statistics
The use of graphs or tables to summarise data
What does standard deviation tell us
How much the scores are spread out from the mean
Evaluate standard deviation
Pro=more precise as includes all scores
Con= distorted by a single value
What does it mean if there is a:
Large standard deviation
Small standard deviation
Large= scores deviate a lot from the mean- very spread out-less consistent
Small- scores don’t deviate much-less spread out-more consistent
When should we use a:
- bar chart
- scatter graph
- histogram
Bar chart=discrete data/clear categories
Scatter graph=correlation
Histogram=continuous data
What does significance mean
A statistical term that indicates the association between two variables is strong enough to accept the ALTERNATIVE HYPOTHESIS
What is the standard level of significance called
5% significance level (p=0.05)
If the results for the statistical test is significant what does this mean
Reject the null hypothesis
Accept the alternative hypothesis
If the results of the statistical test are not significant what does this mean
Accept the null hypothesis
Reject the alternative hypothesis
What is a type 1 error
FALSE POSITIVE
errors which occur when the results are significant but when they actually occurred due to chance
More likely to occur the larger the probability level used
What is a type 2 error
FALSE NEGATIVE
Occurs when results are not significant but they actually were
More Likely to occur the smaller the probability used
What is a level of measurement
Refers to the data gathered off participants
What is nominal data
Data that is in categories and frequencies
What measure of central tendency is used for nominal data
Mode
What is ordinal data
Scores in rank order
Any numerical score
What measure of central tendency is used for ordinal data
Median
What is interval data
Data that has intervals between each score
The intervals are equal
What measure of central tendency is used for interval data
Mean
Why do we use a statistical test
To allow us to make decisions about the result
To allow us to decide objectively whether the results are significant or not using standardised tests
To allow us to accept or reject the correct hypotheses and make conclusions of inferences from our research
Name the test of difference for independent measures for reach type of data
Nominal= chi squared test
Ordinal= Mann Whitney U-test
Interval=unrelated t-test
Name the test of difference for a repeated measures or matched pairs experiment for each type of data
nominal data-Sign test
Ordinal data-Wilcoxon sign test
Interval data-Related T-test
Name the test of relationship for a correlation for each type of data
Nominal- chi-squared test
Ordinal-spearman’s RHO correlation
Interval- Pearson’s correlation coefficient
How can u tell if the results are significant or not
Compare the calculated value with the critical value
Then u decide to accept or reject the null and alternative hypothesis
Name the tests which are GREATER or equal to the critical value
Chi-squared test Unrelated t test Related t test Spearman’s RHO Pearson’s correlation coefficient
What do u need to know to find the critical value
Level of significance
One tailed or two tailed
Number of participants
What is a one tailed/two tailed test
A one tailed hypothesis is directional
A two tailed test is non directional
Define ethical issues
These arise when a conflict exists between participants rights and researchers needs to gain valuable and meaningful findings
What are the 4 ethical issues- briefly describe
1) informed consent- participants must be informed clearly about the purpose of the research so can make an informed decision about whether or not to participate
2) deception- not told true aims of the study- cannot make an informed consent
3) protection from harm- participants should not experience negative physical or psychological effects
4) privacy and confidentiality- concerns participants personal information being protected
How would you deal with:
Informed consent
Deception
A debrief and right to withdraw
How would you deal with protection from harm
Design- researchers to not expose participants to any harm beyond what they might experience in real life
Counselling may be provided
How would you deal with privacy
Anonymity- participants identified as a number rather than their name
Why should we use statistical tests
Allow us to draw conclusions from the results
Allow us to decide objectively if the results are significant
Allow us to accept or reject the correct hypothesis
What are the three different types of statistical test for nominal
Unrelated test of difference- chi-squared test
Related test of difference- sign test
Test of relationship ( correlation )- chi-squared test
What are the different statistical tests for ordinal data
Unrelated test of difference- Mann Whitney u test
Related test of difference- wilcoxon sign test
Test of relationship- spearmans RHO correlation
Describe the different statistical tests for interval data
Unrelated test of difference- unrelated t- test
Related test of difference- related T-test
Test of relationship (correlation)-pearsons correlation coefficient
What does unrelated mean
Independent groups
What does related mean
Repeated measures/ mixed pairs
What is nominal data and what central tendency is used
Data is in categories and frequencies- mode
What is ordinal data and what central tendency
Scores in rank order from a rating scale
Median
What is interval data and what type of central tendency is used
Data which has equal intervals between each score- mean
To decide whether results are significant, what are we comparing
The CALCULATED VALUE from the statistical test
With the
CRITICAL VALUE found in the table
What must happen for the results to be significant
In stats test with an R in there name only
The calculated value must be gReater than the critical value
__________________________________________________
However in stats tests that don’t have an R in them
The calculate value must be lower than the critical value to be significant
What three things must you know to calculate the critical value
1) level of significance
2) once tailed or two tailed
3) number of participants in the study
What does one tailed mean
Directional hypothesis
What does two tailed mean
Non directional hypothesis
define reliability
refers to how consistently it measures what it is supposed to measure
name two ways to asses reliability
test-retest reliability
inter-observer reliability
define test-retest reliability
the experiment is carried out initially and then again at a later date
when the two sets of data are compared, a correlation coefficient of +0.8 demonstrates that the measure has reliability
what is the correlation coefficient that shows the measure has reliability using test-retest reliability
+0.8
what is inter-observer reliability
the extent to which there is agreement between two or more observers
when using inter-observer reliability, what is the correlation coefficient for the results to be considered reliable
it should exceed +0.8
how would you improve reliability when using test-retest reliability
1) reduce ambiguity of questions such as in a questionnaire
2) ensure procedures are exactly the same as before
how would you improve reliability when using inter-observer reliability
1) make behavioural categories clearer
2) practise using the behavioural categories
define validity
the extent to which something is measuring what its supposed to measure ( accuracy)
what are the two types of validity
internal validity ( within the study) external validity (after the study)
what are the four types of validity
1) face validity- seeing if it looks Like it measures what its supposed to measure
2) concurrent validity- comparing a new test with an already well established test
3) ecological validity- the ability to generalise beyond the particular setting
4) temporal validity- ability to generalise research beyond the particular time period of the study
how would you improve face validity
revise and improve the questions to relate more obviously to the topic
how would you improve concurrent validty
remove irrelevant questions
what are the five features of science
Theory hypothesis testing empirical methods (direct testing) objectivity replication
what does falsifiability mean
the possibility that a statement or hypothesis can be disproved
define paradigm
an agreement upon a set of assumptions and methods about the subject matter of a discipline
define economy
the state of a country in terms of the production
what’s the purpose of an abstract
provide a short summary of the study- allows reader to see if report is relevant
what’s the purpose of an introductuon
provides background information on theories and studies relevant to the investigation
explains how hypothesis came about
described past research
whats the purpose of method
provides the specific design details of the study
allows other researchers to replicate the research or potentially improve the method
what’s the purpose of the results section
include any descriptive statistics or any inferential tests
what’s there purpose of discussion in a report
- dsicuss finding sin relation to a previous report
- implications for human behaviour
- outline any limitations
- suggest areas for further research
what’s the purpose of referencing
helps the reader find original source
adds authenticity to the argument
