Research Methods Flashcards
- Aim =
- Hypothesis =
- general statement about purpose of the study
- clear testable statement that states realtonship between variables as stated at the start of the study. It is the prediction.
Three steps to make a strong hypothesis
T - testable > a way of testing the hypothesis
O - operationalised> making sure the variables are measurable, eg; specific measurement like cm or metres or test score
P - precise
EG: participants who drink one can (O) of red bull a dat will stay awake for more hours ( DIRECTIONAL) than people who dont.
Three types of hypothesise
Directional - clear difference anticipated between two conditions/ groups
eg: girls are faster than boys at running
Non- Directional - only states that there is a difference between conditions/ groups.
eg: girls and boys have different levels of running ability.
Null hypothesis - when there is no difference between any condiiton
When to use what hypothesis
Directional: when there are previous studies/theories that suggest a particular outcome\
If there is no previous studies or evidence then jsut non directional
What is an IV
A DV
IV - variable manipulated by researcher
DV - is measured by the researcher
How to test effect of an IV
Two levels
Control condition- no chocolate ( as a comparison )
Experiment condition - yes chocolate
Situational variables
environmental variables, weather, time of day..
Participant variables
participant influences, their intelligence,personality, age
Investigator variables
unwanted influence from the investigator in the reasearchs outcome.
bias, instructions, facial expression, leading questions, smiling, uniform.
Target population
target population - individuals a researcher is interested in experimenting on
Sample population
impossible to practically study everyone in target population so you use a sample of the target population. Sample must be representative so findings can be generalised.
Random sampling
-every member of population have an equal chance of being selected
-Computer generated randomiser
- Numbers from a hat
Opportunity sampling
people who are available to the researcher
Use anyone willing to take part
Student opportunity sampling
Volunteer sampling
self selected
Participants volunteer to participate
Usually through newspaper ads
( Milgram/Zimbardo )
Stratified sampling
Match up sample to target population
How to:
- identify different sub groups that make up population
- worldwide out proportions needed for sample needed to be representative
- participants from sub groups selected through random sampling
Systematic sampling
Every nth number of the target population is selected ( every 6th person in a list of population )
Pros/ cons: volunteer sampling
Pro
- cheap and quick and minimal effort
Cons
- demand characteristics: ( screw u / please you effect )
- low population validity and biased.
Pros/ cons: opportunity sampling
Pros
- cheap and easy
Cons
- low population validity ( usually uni students who volunteer) not representable
- researcher bias, lot of control over who is used in the study
Pros/ cons: random sampling
Pros
- representative and has good population validity
- low researcher bias
Cons
-time consuming
- not always random, can still end up with biased sample
Pros/ cons: stratified sampling
Pros
- representative, shows unbiased sample of population
- more generalisable
Cons
- time consuming
- cant always get a big enough sample
Pros/ cons: systematic sampling
Pros
- reduces researcher bias, no control over who is selected
- representative
Cons
- still possible to get all male sample for eg
Repeated Groups and
Pros/cons
one group of participants that take part in both conditions
Strengths:
- decreased participant variables because its the same people redoing the experiment then there is no variables to affect that ( IQ, eyesight, age, personality ) / demand characteristics/ extraneous variables
- fewer participants (smaller sample size)
CON
Order effects can occur, in repeated groups.
These effects are things such as:
practice effects: once they’ve done the first condition the participants will be well practiced to complete the second one.
Or they could be tired after the first condition, affecting their ability in the second condition.
Order effects are confounding variables.
Repeated measures design
Randomly allocate participants
ABBA
Counterbalance the participants, this reduces order effects. This means 1/2 participants from group do task A then B, and the second half to task B then A
This limited the practice effects
Independent groups
PROS CONS
-two separate groups of participants, one taking part in condition a and the other in condition b
Pros
Decreased order effects
Cons
Increased participant variables
Independent measures design
Cause participants only take part in one condition experimenter has to ensure that participant differences are spread out
Tjhis is through random allocation
Random allocation - randomly select participants so there is less chance of having all the same type of person in one condition.
Matched pairs
Two separate groups
Matched for certain qualities like age or intelligence
One of each pair takes part in condition a adn the other in condition b
Pros
- reduced order effects and reduced participant variables
- reduced demand c
- results therefore more valid
Cons
-more participants are required: increased sample size
- impossible to exactly match each participant
What are pilot studies
A trial run of the investigation before the real one is carried out
What are pilot studies for
this is in order to check the procedure and allow researchers to identify any problems
They are usually run with smaller number of participants
Save time and money that could possibly be wasted in the real thing
How to reduce experimenter bias and demand characteristics in pilot studies
Single Blind procedure: when participants arent told the aim to reduce demand characteristics.
Double Blind procedure: when experimenter and participants dont know the aim of the procedure using a third party in the party to reduce both demand characteristics and experimenter bias.
Control groups: a neutral group who also do the test to compare the experiment group results against.
How to get rid of extraneous (situational) variables
Standardisation - if you make both conditions have the exact same environment then there shouldn’t be any situational variables
Participants are less likely to interpret the study in different ways, reducing extraneous variables
What is an extraneous variable
These are variables other than the IV that MAY affect the DV, variable being measured.
Can be: participant, investigator, situational variables or demand characteristics.
Not systematic: noise, personality, age, temp, time of day.
what is a confounding variable
Variables other than the IV that HAVE affected the DV already and CONFOUNDED the results.
Eg: if an experiment is completed, group A doing it in the morning and B doing it in the afternoon, time would be the confounding variable because it has varied and affected the DV already and is not the iv. It is also something that can be a n extraneous variable and turn into a condfouning variable.
Internal validity
External validity
Internal validity- extent to which the investigation measures what it intends to measure
External validity- extent to which the findings can be applied to real life.
Types of experiments
- Lab
- Field
- Natural
- Quasi
Lab experiment
Conducted in a tightly controlled environment where the IV is manipulated and the researcher observes the effect of this on the DV.
Why does having extraneous/ confounding variables affect the study
Internal validity= measure of accuracy of a study
Having confounding variables lowers the accuracy and therefore validity of a study as we cant be certain the change in the IV caused the effect in the DV
Field experiment
carried out in everyday environment, experimenter still controls IV and extraneous variables are less easy to control.
Natural experiment
Iv is naturally occurring
No manipulation of the IV
Used for more ethically sensitive topics
Natural environment
Quasi experiment
Can be controlled or natural
The difference is that the IV is a pre-existing thing that cannot be manipulated by the researcher, usually differences between people (gender/age)
Pros/cons of lab experiment
Strengths
high levels of control
Cause and effect can be concluded, high internal validity
Reliable and easily replicated
Limitations
artificial therefore low ecological validity cannot be generalised and low external validity ( applicability)
Low mundane realism
Unnatural behaviours displayed therefore demand characteristics can be shown
Pros/cons of field experiments
Strengths
high ecological validity cause environment is more natural
Reduced demand ahracteristscs,experimenter is less obvious
Limitations
lack of control over extraneous variable,
Not reliable
Ethical issues
Pros/cons of natural experiments
Strengths
high ecological validity cause environment is more natural
Can be used to study more ethically sensitive studies
High external validity ( ecological validity )
No demand characteristics
Limitations
participants not randomly allocated to experimental conditions could be a sample bias
No control over IV over extraneous variables
Pros/cons of quasi experiment
Strengths
- lab conditioned- high level of control
Limitations
- bias in the procedure because you cannot use random allocation to sort participants
- no control over environment or extraneous variables like age or gender and cant ensure they wont affect the dv
- lab: lack of ecological validity
- findings cannot be generalised to real life setting
Whats an observation
Observation: watching and recording peoples behaviour usually in a natural environment.
helps psychologists see natural behaviour in that natural setting, a more realistic conclusion.
2 types of observations based on presence of researcher
- participant and non participant observations
Observations based on environment of observation
- naturalistic and controlled
Observations based on awareness of participants in observations
Overt and covert observations
What are participants and non participant observations
Participant observations:
the observer acts as a part of the group being watched
Very subjective observation because you experience it as well
Non participant observation
observer is not in the group
Behaviour is recorder objectively
More practical and ethical
What are naturalistic adn controlled observations
Naturalistic observation:
observations take place in a natural setting to observe natural behaviour
Natural/quasi experiment
Controlled observation:
some variables are controlled and manipulated
Still manipulation over IV and attempt it control extraneous variables
What are overt and covert observations
Overt observations:
open observations
Participants are aware of being observed and the aims of the study
Gave consent to be observed
Covert observations:
unaware of observations and havent given consent
Usually observation of public behaviours
4 different types of questions in questionnaires
Open
Closed
Fixed qnswers
Rating scale
6 strengths of questionnaires
SIQQ
S- large sample size can be gathered which might be less obtainable in an experiment
I - investigator bias is low as they can be completed without researcher
Q - qualitative data
Q - quantitative data
Two cons of using questionnaires
LeSS
Le - leading questions, encouraging a certain answer from participants is biased
S - superficial
S- social desirability bias
Open quetionnaires
- open questions that provide qualitative data which includes subjective information, rich detail but might be harder to analyse
Can be answered in any manner no limits
Closed questionnaires.
Has a fixed response ‘yes or no’
Or a rating scale, likert scale etc
These provide quantitative data - easy to analyse but might lack meaning and detail