research methods Flashcards
sample
participants selected and used in the research that are a part of the larger group (population) or a subset
population
- entire group of research interest
- the larger group
- a sample is selected from
- researcher apply (generalise) the results
sampling procedures
random sampling
random stratified sampling
stratified sampling
random sampling
ensures every member of the population has an equal chance of being selected to be part of the sample
how to obtain random sample
obtain a sampling frame e.g electoral roll
use simple ‘lottery procedure’ e.g names out of a hat
limitations of random sampling
the sample may not be representative (bc need to use law of large numbers)
random stratified sampling
ensures the sample is highly representative of the population= therefore not biased. participant randomly selected from each sub groups (strata)
stratified sampling
dividing population into different subgroups e.g gender, age, intelligence
limitations of stratified/random stratified sampling
need a complete list of the targets populations
time consuming
method to obtain a stratified sample
- the population is broken into differnt groups based on a variable you want to control e.g age
- individuals= selected from each group- usually in the same proportion that they appear in the population (e.g if pop. contain 20% year 9 students, then sample= include 20% of year 9 students.)
- ensures characteristics of the poulation= captured leading to a more representative sample
participant allocation
- random allocation
- counterbalancing
random allocation
- less likely to be biased but risk of indiv. diff.
-every participant has an equal chance of being selected for any of the groups used (experimental vs control)
counterbalancing
-use to minimise/ balance out order/ practice effect
- involves systematically changing order of treatments/tasks for p;articipants in a balanced way to counter the unwanted effects of order effect
experimental research
used to assess a cause-and-effect relationship between variables under controlled conditions
controlled variables
varibale= held constant in an investigation.
ensure the manipulation of the IV causes the changes in the DV
independent variable
variables that are manipulated
assumed to have a direct effect on the dv
dependent variable
varibale that is measured
extraneous variable
any varibales other than the iv that has been identified and controlled for
confounding variables
any variables other than the iv that has not been identified and controlled for
experimental group
group exposed to experimetal condition (iv)
control group
group that is exposed to control condition (iv is absent)
why operationalise
ensure IV and DV are testable
Helps to avoid experimenter bias
allows possible reptition of the experiment
hypothesis
It is hypothesised that those who are (exposed to IV) will (impact on DV) in comparison to those who have not (exposed to IV)
random error
unsystematic and occur due to chance
systematic error
consistently favours one condition over another (in-built flaw) hence errors in data that differ from the value by a consistent amount
within subjects design
the same participants are used in both experimental and control group
limitation of within subject design
- order/practice effects = participants may perform better the second time around due to practice
(control using counterbalancing) - fatigue effect = participant may perform worse the second time around due to fatigue or boredom
- high drop out rates
between subjects design
participants randomly allocated to one of two (or more) entirely separate groups
strength of w/in sub design
- less participants required
- eliminates differences between groups as same participants are used in both groups
limitation of b/w sub design
-participants are not matched on characteristics (indiv diff)
- eed more participants compared to repeated measures
strenght of b/w sub design
- avoid order/practice effect and fatigue effectb/c each participant only does experiment once
- less time consuming
mixed design
elements of a b/w sub and w/in sub design are combined
why use mixed design
researcer examines not only the potential differences between 2 or more separate groups of participants but also assesses change in the individual members of each group over time
strength of mixed design
- differences in participant variables b/w groups= controlled in the w/in sub design element
- can test the effect of multiple on a DV in one investigation
limitation of mixed design
- higher participant withdrawal rate from stud
- can effect internal validity
- less control over participant knowledge of the study (prior participation in the first condition may influence behaviour in the second condition
- less control over differences in participant variables in the b/w sub element
what does systematic error affect
accuracy
what does random error affect
precision
repeatability
produce the same results when carried out under identical conditions within a short period of time e.g same procedure, observer, instructions etc
reproducibility
produce the same results when repeated under different conditions
what needs to be considered when writing conclusion
the data (mean and SD)
what impact would EV and CV have on DV
how to conclude conclusion
However, due to lack of inferential statistics, it is unclear the extent to which the results are due to error
validity
refers to the extent to which procedures used for a research study measures what it intended to measure
internal validity
the results obtained for a study are actually due to the variable(s) that was tested or measured and not some other factor (EV)
external validity
the results obtained for a study can be generalised to the target population
ethical concepts list
(can ben jump in, not run?)
The Concepts are:
Beneficence
Justice
Integrity
Non-maleficence
Respect
ethical guidelines list
(can voldemort win if dembledore dies)
Confidentiality
voluntary partcipation
withdrawal rights
informed consent
deception
debriefing
informed consent
needs to be signed-written contract and if minor or vulnerable, need to get consent from the parent. guardian and themselves
four factors relevant to evaluating psychological issues, research and practice
- sociocultural factors
- economic factors
- political factors
- legal factors
sociocultural factors
environmental conditions that impact the practices, beliefs, social norms and expectations of individuals or groups
e.g family support, education history and opportunity, and availibility of healthcare
economic factors
financial factors e.g income of individuals, financial characteristics of a study e.g funding of research
political factors
the environmental conditions that impact the beliefs and actions of groups and individuals, groups, and organisations
non-experimental research studies
(Christopher Charles Calleja Loves Playing Fairies & Singing Material-girl )
Case study
Classification&identification
Correlational study
Literature review
Product, process or system development
Fieldwork
Simulation
Modelling
case study
intensive, in-depth investigation of some phenomena of interest in an individual, group, organisation or situation
strength of case study
- provide highly detailed, rich information about a particular phenomenon under study
- provide new knowledge about other phenomena
- can incorporate other scientific methodologies to gain data
limitation of case study
- results cannot be generalised to a wider population as case studies often only involve a small group of people or one person
- time consuming
- subject to researcher bias and errors, as often conducted by one/only a few researchers
classification and indentification
classification:
arrangement of phenomena, objecs or events into manageable sets
Identification:
assigning certain things to their respective label/group
strength of class. and identif.
-allows scientists to form more targeted solutions/interventiosn to real problems
- provides a common language to communication about scientific phenomena (fact/situation that is observed to exist/happen)
limitations of class. and identif.
-labels and language can be inaccurate and create bias
- over-simplify reality
correlational studies
researchers observe and measure the relatiosnhip between 2 or more variables without any active control or manipulation of them
correlational vs causation
correlation:
refers to strength of relationship b/w variables e.g how likely it is that they would occur together in some predictable way
Causation:
refers to a relationship between variables = change in one variable causes a change in another e.g hot weather causes sunburn and an increase in ice cream sales
strengths of correlational studies
no manipulation of variables required
limitation of correlational studies
- can be subject to the influence of extraneous variables
- results cannot draw conclusions about cause and effect
fieldwork
research involving observation and interaction with people and environments in real-world settings, conducted beyond the laboratory
- data is collected in the real-world, authentic setting
strength of fieldwork
can be conducted in naturalistic settings hence finding more applicable to the real world
limiations of fieldwork
- cannot manipulate environment
- can be time-consuming and expensive to conduct and then record data
- difficult to replicate b/c long procedures in a real world setting
literature review
collating and analysing secondary data related to other people’s scientific findings and or viewpoints
strength of literature review
may uncover patterns of knowledge/gaps of knowledge
limitations of lit review
- may be difficult to do if little research has been done on a topic
- time consuming
modelling
manipulation of a physical model
e.g a small/large-scale representation of an object/ a conceptual model that represents a system to help people understand, know/simulate the system
conceptual model e.g
e.g multi-store model of memory
physical model e.g
plastic human brain, plastic eye
strength of modelling
- can provide explanatory tools
- allows researchers to know, understand and problem solve (physical model)
limitations
- as models are used to simplify and communcate ideas, they may over simplify or inaccurately represent reality
product/process/system development
created on the basis of scientific research and development that help people in their daily life
strength of product etc
creates products/ processes and systems that may meet a human need
limitation of product etc
can be expensive and time-consuming
simulation
using a model to study the behaviour of a real/ theoretical system
strength of simulation
- provides insight into potential circumstances and events
- allows research to view micro, hard-to-see phenomena e.g neuron in detail
limitations of simulation
- time-consuming and expensive
- subject to programming and human error so may not always be an accurate prediction/reflection of reality
