Research Methods Flashcards
What is an aim
A statement of what the researcher intends to find out
What is a hypothesis
Precise and testable statement specifically of the relationship between variables
What is a directional hypothesis
A statement that states the kind of difference very between the two conditions or participants
What is a non directional hypothesis
A statement that simply states there will be a difference between the two groups
When would we use each hypothesis
Directional- when there is previous evidence known the researcher can predict the direction of the findings
Non directional- when there is little precious evidence or evidence is contradictory
When do we use experimental hypothesis and research hypothesis
Experimental in experiments and research for everything else
What is a null hypothesis
A statement of no difference
What is the alternate hypothesis
It is alternative to the null, says something will happen
Which do we choose from
We write both then reject one
What must all hypothesis include
Both variables and it must be operationalised
What are the levels of IV
The IV needs to be compared to a different level eg loud music and no music
What’s an extraneous variable
Any variable other than the IV that can can affect the DV if it’s not controlled at the start of the experiment
What is a confounding variable
An uncontrolled extraneous variable that negatively affects results at the end of the experiment
What is a participant variable
Things about the people in your study
What is a Situational variable
How the environment affects the behaviour
What is an experimenter variable
Things to do with the person conducting the experiment
What is social desirability
We want to be socially desired so we answer accordingly
What is social desirability a form of
Demand characteristics
What is demand characteristics
Changing your behaviour on purpose to affect the results
Name the 3 experimental designs
Repeated measure, independent group, matched pairs
What is repeated measure
Repeating conditions with all participants
What is Independent groups
Divide groups into 2 and do separate conditions
What is matched pairs
Two groups, but each participant must have a pair in the opposite group with a matched, relevant variable
Evaluate repeated measure
There is participant variability and less people needed, but there are order effects and demand characteristics
Evaluate independent groups
Demand characteristics and Oder effects are removed, but there is more participant variability
Evaluate matched pairs
There is less participant variability and order effects and demand characteristics are removed but there are practical issues
Name the hypothesis framework
“Participants who (IV1) will (DV) more/less than participants who (IV2) “
Name the 4 types of experiment
Lab, field, quasi and natural
Explain lab experiments
A highly controlled environment in which the researcher manipulated the IV and records the effect on the DV
Explain field experiments
Same as lab but takes place in a natural, more everyday setting
Explain a quasi experiment
It’s almost an experiment, as the IV is an existing difference between people e.g gender
Explain natural experiments
When the researcher takes advantage of a naturally occurring IV
What is random sampling
Where you take a group of people, assign numbers and use a random number generator
Evaluate random sampling
Unlikely that it will be biased and everyone has an equal chance of being chosen, but it can take a long time
What is stratified sampling
The sample is a proportional representation of the target population by breaking it down into smaller groups and picking one from each
Evaluate stratified sampling
It is very representative and equal, but it is very time consuming and difficult
What is the target population
The people you are aiming to generalise the findings to
What is opportunity sampling
The sampler selects people who are available at the time
Evaluate opportunity sampling
It is convenient, easy, quick and ethical but can be less representative
What is systematic sampling
Gaining samples by choosing every 5th, 7th and 11th person on a register
Evaluate systematic sampling
Avoids researcher bias, but it is not representative/ can be faulty eg you get an all male sample
What is volunteer sampling
Participants putting themselves forward for the experiment
Evaluate volunteer sampling
Participants would be motivated, they would have given consent and you have access to many people but all volunteers are likely to be of the same mindset and therefore not representative
The bigger the sample to size the what
More representative therefore the easier it it to apply your findings
What are ethics
Standards about what is right and wrong
The conflict between what a researcher wants to do and the rights of participants
What are the types of ethics
Informed consent Deception Confidentiality Debriefing Right to withdraw Protection from harm
Who controls the ethics
BPS
What is a pilot study
Small scale trial run
What are pilot studies used with
Questionnaires or interviews
What do pilot studies allow researchers do
Check coding systems in observational studies
Identify any problems
Gives time to modify the design, procedures and resources
Does it collect any data
No
What are the two types of pilot studies
Single blind and double blind
What is a single blind procedure
The aim is specific but the researcher doesn’t tell the participants
Participants placed in either experimental or control and this is done to prevent demand characteristics
What is a double blind procedure
Neither the participant nor the researcher knows who’s in what condition
Only lead researcher knows
This stops researchers acting differently
What two control groups are there
Real drug - experimental
Placebo - control
What is an unstructured observation
Observation with no structure or aim, provides rich info
What is a positive of an unstructured observation
Provides rich information
What is a negative about unstructured observations
Researchers tend to record the most eye catching things which is highly subjective
What type of data is produced with unstructured observations
Qualitative
What is a structured observation
When we break behaviours down into categories that are precise and measurable
What must all behaviours be in the structured observation
Operationalised
Define inter-rater reliability
The extent to which all observers see and rate the same behaviours during an observation
How can we increase inter-rater reliability
Reduce differences by training
When do problems arise with categories
When they aren’t mutually exclusive and behaviours fit into more than one category
What is observer bias
An observer can see things through their own way
What can we do to check correlation
Use statistical tests to check it is higher than 80%
What type of data do we get from structured observation
Quantitive data
What are sampling methods
How to record data in observations
What are two things we can base our observation categories on
Recording data at pre determined times
Counting the number of times an event occurs
If the words correlation, relationship or association are in a graph question, which graph should you use
scatter graph
If there is a difference between categories, which graph should you use
bar chart
What is the layout for the title of a graph
A _____ to show the difference/relationship between (operationalised variables)
What must we operationalise with a graph
all labels of axis and title
Name two types of discrete data
gender, different conditions
Name two synonyms for discrete data
Nominal/non-continuous
Name two types of continuous data
Age, height
What does a normal distribution look like
symmetrical and a bell curve shape
What occupies the midpoint of a normal distribution
mean, median and mode
What is a skewed distribution
when they lean to one side or another
What is a negative skew
Most distribution is on the right side of the graph with a long left tail
What is a positive skew
most distribution is on the left side with a long right tail
What causes a skew
a few extreme outlying data points
Where is the mode on a skewed graph
the highest point
Which was from the mode do the median and mean go
towards the tail
Which is most affected by the skew
the mean
In a perfect normal distribution, what percentage of the population is between the mean and 1SD either above or below it
34.1%
What is the standard probability in psychology
5% chance aka P=0.05
When might we choose a stricter level of significance
when we need more confidence in the results being accurate such as human medicine trials
What is the probability used in human medical trials
1% P=0.01
When should we use the sign test
When we are looking for a difference not an association
when a repeated measures design is used or matched pairs
when data is in categories
What 4 things do you need for the sign test
table of critical value
number of participants
significance value (0.05)
directional or non directional hypothesis
Why are statistical tests used
Statistical testing provides a way of determining whether hypotheses should be accepted or rejected, and they tell us whether differences or relationships between variables happen by chance or are statistically significant
What is a causal relationship
the cause and effect of IV on DV
When only can a causal relationship be established
in a true experiment in a lab/field
What is peer review used for
used to ensure any research is of high quality, genuine and trustworthy before being published
what is the process of peer review
Sent to editor of a journal
This is then checked by a panel of 2/3 experts
Methodology, analysis and conclusions are scrutinised
What are the three main outcomes of peer review
Research is approved for publication
Sent back for amendments
Rejected
What are the three main purposes of peer review
check quality so it can be trusted by the general public
ensure accuracy of findings before publication
to allocate findings
What are the three ways in which a peer review can be carried out
single blind, double blind and open review
Explain each
Open review – the researcher knows who is reviewing them and the reviewers know who they are reviewing
Single blind – the researcher does not know who is reviewing them but, the reviewer knows who they are reviewing
Double blind –the researcher does not know who they are reviewing and, the researcher does not know who is reviewing them
What are the strengths of anonmity in reviewing
Anonymous reviewers can be honest
Anonymous researchers won’t be overly criticised by rival peers
What are the limitations of anonmity in reviewing
A reviewer might reject or delay approval until after their own research has been published
A reviewer might reject someone who has rejected them in the past
They might reject a proposal so they can be allocated funding
Explain the two evaluation points for peer review
publication bias - Academic journals tend to publish positive results (file draw problem), they post the stories that will sell
burying ground breaking research - Ground-breaking research may be buried if it challenges popular ideas or influential researchers
What does psychology in the economy concern
production, consumption and distribution of goods
What is the behavioural insights team also known as
the nudge unit
What does the nudge unit do
an organisation set up to apply behavioural economics to try and improve government policy and to save the UK government money.
What type of social influence does it involve
normative as people dont want to be the minority
Why is helping peoples mental health good for the economy
1/3 of absences are due to mental health and absences scost the economy £15 billion a year
What do effective treatments for mental health mean for the work force
sufferers are able to manage their condition and return to work reducing absenteeism
be more productive at work (presenteeism)
reduce burden on social services and cost to NHS
Explain the economic implications of attachment research
role of father is now viewed as equally important, meaning the if the mother is a higher wage earner then she can go back to work
Explain the economic implications of memory research
If EWT is inaccurate, this means a delayed and flawed justice system which ultimately costs the state
What is reliability
when you get the same consistent result every time
How did we used to measure IQ
head circumference
What is internal reliability
consistency within a test
What is external reliability
produces similar results on each replication
What is inter-rater reliability
different raters produce similar findings
What is the test-retest method
repeat the test with the same people on different occasions
When is this most commonly used
questionnaires and psychological tests
What type of reliability does this test
external
What test is used to test internal reliability
split half test
Explain the split half test
test split in half and same participant does both halves, should get same results
What causes subjectivity bias
when a researchers view of events may be fundamentally different
Name 3 ways we can improve reliability
take more than one measure
pilot studies
standardise the way data is collected and train observers
How can we maximise internal validity
by maximising control over extraneous variables
What does maximising control over extraneous variables ensure
that the course and effect is correct
What are the two types of external validity
population and ecological
What is population validity
can the results be generalised to the population
What is ecological validity
can the results be generalised to the real world
How can we make ecological validity high
the tasks must involve mundane realism (everyday things)
What is temporal validity
do findings from a study remain true and valid over a period of time
What is face validity
does it measure what you are wanting it to measure
What are the 4 types of validity that you should talk about in an exam
internal (whether the methods are valid)
population
ecological (is it natural)
temporal (has it been repeated frequently and recently)
Name two reasons why we use statistical tests
work out the probability of whether a particular set of data could have occurred by chance
we can see if a difference between behaviour is statistically significant
What does H1 and H0 mean
h1 - hypothesis
h0 - null hypothesis
What are the three D’s used for
used to choose which stat test to use
Name the three D’s
difference or correlation
design - related (repeated measures) or unrelated (independent measures)
Data - what type
How can we find out if its a difference or correlation
look at title - association between two variables (C) or IV/DV (D)
Which design is unrelated and why
independant groups as there are different people/participant variables in each condition
Which designs are related
matched pairs and repeated measures
What is nominal data
categorical data that is grouped
What is ordinal data
data that is ranked or ordered
What is interval data
public measurement of something real eg time, heart rate, speed
Which measure of central tendency relates to each data type
nominal - mode
ordinal - median, range
interval - mean, standard deviation
When would you use a parametric test
data from a normal distribution
variance of two groups should not be significant
What is the mnemonic for the stat test table
carrots should come mashed with swede under roast potatoes chi squared sign test chi squared mann whitney wilcoxon spearmans rank unrelated t test related t test Pearsons r
What does probability mean
the likelihood of something happening
What is the probability equation
p = number of particular outcomes divided by number of probable outcomes
What does 0 and 1 mean in probability
0 = statistically impossible 1 = statistical certainty
What does significance mean
the level to which the null hypothesis is accepted or rejected
What is the role of probability and stats tests
to be able to accurately reflect what would happen in the population
What is a type 1 error
When null is rejected and should have been accepted
false positive
What is a type 2 error
When null is accepted when should have been rejected
false negative
When do we find these errors
type 1 - too lenient with significance level
type 2 - too stringent with significance level
Define a case study
an in depth investigation of a single individual or group
Define content analysis
a research technique that enables the indirect study of behaviour by examining communication through texts, emails etc
Define coding
identifying chosen categories used in content analysis
Define thematic analysis
qualitative approach to analyse implicit or explicit meanings in content
Name some strengths with case studies
rich, detailed insights
longitudinal so can see any changes
contribute to understanding of normal functioning
Name some weaknesses with case studies
often atypical cases that cant be generalised
rely on first hand accounts from family that may remember things incorrectly
Name some strengths with content analysis
an get around ethical issues as no consent needed
high internal validity as no demand characteristics
provides quantative and qualitative data
Name some weaknesses with content analysis
researcher may attribute thoughts or feelings to a text that weren’t meant by the writer
lack of objectivity
What is a feauture that makes psych scientific
empirical evidence
What is empirical evidence
evidence that is directly observable and measurable
For a method to be considered empirical what must it also be
objective
What is objectivity
methods are not influenced or biased by beliefs and opinions of observer
What is needed in a method to ensure objectivity
research must maintain a critical distance
What else must research be in order to be scientific
replicable
What did Popper say about falsifiability
scientific theory must be at least possible to disprove
What is a paradigm
a theory that everyone accepts eg theory of evolution
What do the methods need to be open to in order to be falsified
hypothesis testing
How to write a design a study 12 markers
address each bullet point in turn be specific - number of p's, design, timings, sampling and why? details of tasks and materials step by step guide ethics variables
What is presenteeism
working whilst exhausted
Once research is complete, what is done
a report is written, following a prescribed format with a number of sections
What are the 6 sections
abstract, inroduction, method, results, discussion, referencing
Explain the abstract
150-200 word summary
written last
access to abstract free so can let people decide whether they want to purchase full paper
Explain the introduction
provides reader context by reviewing existing literature
shows why further research was decided to be carried out
broad to begin with then narrows
ends with specific aims and hypothesis
Explain the method
divided into sub headings
enough detail for replicability to establish reliability
design/method, sample, materials, procdure, ethics
Explain the results
no raw data measures of central tendancy summary table graphical displays stats tests hypothesis accept/reject
Explain the discussion
a verbal description of results and conclusion
how findings link to other things
limitations addressed
suggestion of future research
Explain the reference
acknowledgement of sources
name and date of other research
full referencing comes at the end in the references section
