1.2 Quantitative Research - The Experiment Flashcards
What must you do to get the most accurate results?
In order to get the most accurate results, you must isolate IV and manipulate it so nothing else changes
What is a confounding variable?
any unwanted variables that may influence the dependent variable (distort relationship between IV and DV)
What are confounding variables linked to?
linked to/contribute to bias
What needs to be done to confounding variables to reduce the chance of them affecting the results?
Need to be either completely eliminated or kept constant in all groups so they do not affect comparison
What does the experiment aim to do?
Experiment aims to discover universal laws of behaviour applicable to large groups of people across a variety of situations
What is the target popualtion?
group of people to which the findings of the study are expected to be generalised
What is representativeness?
a key property of a sample ensures that info obtained from the sample can be generalised to the target population
Is there a quantitative way to establish representativeness?
NO
How do researchers know which characteristics they want in a sample?
Researchers decide what characteristics are essential
Done on basis of prior knowledge, theories and studies
How is the sampling technique decided upon?
available resources
aim of the research
nature of the target population
What is an advantage of random sampling?
The ideal approach to make a sample representative
Allows all random characteristics to be sampled, making results easily generalisable and every member of the target population has an equal chance of becoming part of the sample
What is a disadvantage of random sampling?
Not always possible for practical reasons:
E.g if based in Europe can not get a sample of every race of teenager from every country
In this case, you either believe that cross-cultural differences are not essential or narrow down your target population
How does stratified sampling work?
More theory-driven
First, decide characteristics that the sample has to reflect
Study distribution of this characteristic in target population e.g use statistical data
Recruit participants in the same proportion as they are in population
How does convenience sampling work?
Recruit participants that are easily available
What are the advantages of convenience sampling?
Easy choice when financial resources and time are limited
Certain studies do not need cross-cultural samples etc
Useful when the wide generalisation of findings is not the primary goal of research e.g only doing an exploratory study and only want to check hypothesis or testing someone else’s theory
What is a disadvantage of convenience sampling?
lack of representativeness
What is self-selected sampling?
Refers to recruiting volunteers
e.g advertising in a newspaper
What is an advantage of self-selected sampling?
Quick and easy + wide coverage
What is a disadvantage of self-selected sampling?
Limit is lack of representativeness e.g people may be motivated than the general population or looking for incentives
What does experimental design mean?
the organisation of groups and conditions in an experiment
What are the 3 main types of quantitative experimental designs?
- independent measures design
- matched pairs design
- repeated measures design
What is the general process of independent measures design?
Involves random allocation of participants into groups and a comparison between these groups.
Randomly assign people to groups i.e control and experimental and manipulate experimental conditions, then compare DV
Why is random allocation used in independent measures design?
The rationale behind random group allocation is that all potential confounding variables cancel each other out
The larger the group higher the chance of the groups being equivalent
When is an experiment said to follow an independent measures design?
The experiment follows an independent measures design when the IV is manipulated by randomly allocating participants into groups, allowing us to assume that groups are equivalent, so whatever difference we observe is a result of the experimental manipulation
How does a matched pairs design look like?
Similiar to above, only difference is that instead of completely random allocation, researches use matching to form the groups
How are people separated into groups in matched pair designs?
Rank people by factor e.g memory, and go from top of list splitting pairs ( 1 person in control other in experimental)
What is the matching variable?
Variable that is controlled e.g memory (thing tested to rank people) is called the matching variable
When is a matched pairs design preferred?
Equivalence in a particular variable is particularly important
The sample size is not large and not sufficient to ensure group equivalence through random sampling
When is repeated measures design used?
Used when the goal is to compare conditions rather than groups of participants
What happens during repeated measures design?
Group of participants exposed to conditions and then conditions are compared
What is the main weakness of repeated measures design?
Vulnerable to order effects - results may be different depending on which condition comes first
What are 2 reasons why order effects may occur?
Practise - participants become more comfortable and confident in the second trial - performance increases
Fatigue - participants get tired during the first trial and their concentration decreases in the 2nd trial - performance decreases
How may order effects be overcome?
To overcome order effects researchers use counterbalancing
What does counterbalancing involve?
Involves using other groups of participants where the order of conditions is reversed
What is an advantage of repeated measures design? (2)
The advantage is that people are compared to themselves - overcomes
participant variability - (the difference between groups before the experiment starts) making the comparison more reliable
Smaller group sizes are required
How is the quality of an experiment measured? (validity) - 3
internal construct external (validity)
What does construct validity evaluate?
Characterises the quality of operationalizations
When is construct validity high?
Construct validity is high when leap from operationalisation to construct is justified and sufficient coverage of construct
Evaluates how an abstract concept can be measured - create an operationalization
Why is operationalisation useful?
makes empirical research possible
When is operationalisation high?
How clear is the link between the abstract thing being measured and the construct
What does internal validity evaluate?
Characteristics the methodological quality of the experiment
When is internal validity high?
Internal validity is high when confounding variables have been controlled and we are quite certain the change in DV is the result of IV manipulation
What is low when internal validity is high?
BIAS
What does external validity evaluate?
Characterises generalizability of findings in the experiment
What are the 2 types of external validity?
population and ecological validity
What does population validity evaluate?
refers to extent that findings can be generalised from the sample to target population
When is population validity high?
Is high when the sample is representative of the target population and an appropriate sampling technique is used
When is ecological validity high?
(when surroundings are more naturalistic) refers to the extent to which findings can be generalised from the experiment to other settings or situations.
What two validities are linked and what is their relationship?
There is an inverse relationship between internal validity and ecological validity
How does the relationship between ecological and internal validity work?
Internal validity grows when we have more control, so conditions are more artificial,
Ecological validity grows when conditions are more naturalistic (less controlled)
What are 9 types of bias for quantitative research methods?
- selection
- history
- maturation
- testing effect
- instrumentation
- regression to the mean
- experimental mortality
- demand characteristics
- experimenter bias
When does selection bias occur?
Occurs if groups are not equivalent at the start of the experiment
How does selection bias affect the results
Cannot be sure if post-experiment differences between groups reflect the influence of IV or other variables
IN what types of experimental designs is selection bias most common?
happens in independent measures and matched pairs design
What is history bias?
Refers to outside events that happen to participants in the course of the experiment
i.e non-natural events
When is history bias particularly intrusive?
In lengthy experiments
How could you counteract history bias?
To counteract history as a threat to internal validity confounding variables should be eliminated or kept constant in both groups
What is maturation bias?
In the course of an experiment, participants go through natural development processes e.g fatigue, growth
What is testing effect bias?
First measurement of the DV may affect the subsequent measurements, as participants can be more familiar with the format/experiment
What type of experimental design is a special type of testing effect bias common in?
In repeated measure designs, testing effect is a special case of order effects with counterbalancing being used to control it
What is instrumentation bias?
Occurs when the instrument measuring the DV changes slightly between measurements
What instrument of measurement is most commonly used in psychology?
In psychology instrument of measurement is usually a human observer
How could instrumentation bias be reduced?
To avoid this researchers should try to standardise measurement conditions as much as possible across all comparison groups and observers
When is regression to the mean common?
Becomes a concern when the initial score on the DV is extreme (low/high)
What is regression to the mean?
If we have an extreme results, subsequent results (by probability) will be more average e.g if extreme anxiety levels they will lower, as extreme results are not probable
How can regression to the mean be reduced?
A countermeasure is a control group with the same starting level/measurements at the same point of time
What is experimental mortality?
Refers to the fact that some participants drop out during an experiment, becoming a problem if dropouts are not random
Why is experimental mortality a problem?
Groups may not be equivalent after large number of dropouts in one of them
What is a way in which experimental mortality can be reduced?
No reliable way to prevent this, only to design experimental conditions in such a way that participants do not feel the need to drop out
When do demand characteristics appear?
Refers to a situation when participants understand the purpose of the experiment and change their behaviour to subconsciously fill that interpretation
What is the bias linked to demand characteristics?
social desirability bias
How can social desirability bias be counteracted?
Only true way to counteract is deception (ethical problem) but also a post-experimental questionnaire to find out to what extent demand characteristics may have influenced results
What type of experiment is social desirability bias most likely to occur in?
repeated measures design -participants may guess aim of experiment and try to sabotage experiment
What is experimenter bias?
Refers to situations in which the researcher unintentionally exerts an influence on the results of the study
how could you counteract experimenter bias?
Counter-measure is using double-blind designs - information that could introduce bias is withheld both from the participants and from the people conducting the experiment
What are the 4 main types of experiments?
Quasi
True
Field
Natural
What is the difference between quasi and true experiments?
Quasi-experiments different from true experiments in that the allocation into groups is NOT done randomly
How are groups decided in quasi-experiments?
The pre-existing inter-group difference is used e.g gender, age, cultural background, occupation
What is a major limitation of quasi-experiments?
Major limitation is that cause-and-effect inferences cannot be made
Cannot be sure of equivalence
Why is a quasi-experiment sometimes preferred over a true experiment?
Sometimes a “true” experiment cannot be conducted as IV CANNOT be manipulated e.g age
How are quasi-experiments similar to true experiments?
Quasi experiments are designed similarly to true studies (superficially)
What are field experiments?
Field experiments are conducted in a real-life setting
Can the researcher manipulate the IV in a field experiments?
Researcher manipulates IV, but since participants are in a natural setting, many extraneous variables cannot be controlled
What is an advanatge of field experiments?
- high ecological validity
What is a disadvantage of field experiments?
Limitation is less control over potentially confounding variables - LOWER INTERNAL VALIDITY
What is the difference between natural and field experiments?
Natural experiments are conducted in a participants’ natural environment but here the researcher has no control over the IV - it occurs NATURALLY
What are advantages of natural experiments?
Ecological validity is an advantage (but internal validity is a disadvantage due to less control over confounding variables)
The advantage is that they can be used when it is unethical to manipulate the IV
