Research methods Flashcards
What is a “variable” ?
(Experimental methods)
Any thing that can vary or change in an investigation.
What is the independent variable (IV) in psychology?
(Experimental methods)
Some aspect of the experimental situation that is manipulated by the researcher - or changes naturally - so the effect on the DV can be measured.
What is the dependent variable?
(Experimental methods)
A variable that is measured by the researcher.
Any impact on the DV should be caused by the IV.
Define operationalization.
(Experimental methods)
Clearly defining variables in terms of how they can be measured.
What is an extraneous variable?
(Research issues)
Any variable, other than the IV that may impact the dependent variable if not controlled.
(Nuisance variables that do not vary systematically with the IV)
What is an experimental method?
Involves the manipulation of an independent variable to measure the effect on the dependant variable.
When do you use a directional or non directional hypothesis?
- A directional hypothesis is where a theory or the findings of previous research studies suggest a particular outcome.
- A non directional hypothesis is used when there is no theory or if previous research findings from past studies are contradictory.
How do you write clear hypotheses?
- Make sure the IV AND DV are clear and measurable.
- You have stated the relationship between the IV and DV, not stating an aim.
- Selected the appropriate hypothesis (ie- directional or non directional)
Why are variables controlled?
Other variables that affect the DV should remain constant, this is because the researcher needs to feel confident that any change in the DV was due to the IV alone.
How would someone operationalise variables?
- Make the variables as measurable as possible, making it testable.
- For example, in ‘speed up’ a researcher can hypothesise that: After drinking 300ml of SpeedUpp, participants say more words in a five minute period than participants who drink 300ml of water.
What is the difference between an aim and a hypothesis?
An aim is a general statement that describes the purpose of the investigation, developed from a theory.
A hypothesis is a statement made at the start of the study that clearly describes the relationship between variables.
What are the 4 types of experimental method experiments?
- Laboratory
- Field
- Natural experiment
- Quasi experiment
What is the difference between a directional and non-directional hypothesis?
- A directional hypothesis is where the researcher makes clear the sort of difference anticipated between the two conditions or groups - usually including words such as more or less, higher or lower or faster or slower.
- A non directional hypothesis simply states that there is a difference between conditions or groups, HOWEVER, the nature of the difference is not specified.
How would you write a hypothesis - CHECKLIST.
- Make sure the IV and DV are clear and measurable
- State the relationship between IV and DV (do not write an aim)
- Select DIRECTIONAL or NON-DIRECTIONAL hypothesis
- Operationalize the variables.
What are the 4 issues in research?
- Extraneous variables
- Confounding variables
- Demand characteristics
- Investigator effects.
What is an extraneous variable?
- Any variable, other than the independent variable (IV), that may affect the dependant variable (DV) if not controlled.
- They are nuisance variables that DO NOT vary systematically with the IV.
What is a confounding variable?
- A type of extraneous variable.
- The main feature is that a confounding variable varies systematically with the IV.
- This means that we can’t tell that any change in the DV is due to the IV or the confounding variable.
What is a demand characteristic?
Any cue from the researcher or the situation that may reveal the purpose of the investigation to the researcher.
This can result in the participant altering their behavior within the research situation.
What are investigator effects?
- Any effect of the investigators behaviour (conscious or unconscious) on the research outcome (DV).
- This can include anything from the design of the study to the selection of, and interaction with the participants.
What is the difference between extraneous variables and confounding variables.
A confounding variable is a type of extraneous variable that varies systematically with the independent variable, whereas extraneous variables generally do not systematically vary with the IV.
What is the difference between demand characteristics and investigator effects?
What are two ways demand characteristics can change participants behaviors in?
- Please-U: Acting the way they believe is expected and overperforming.
- Screw-U: Deliberately underperforming to sabotage the results.
In either case, the participant behavior is no longer natural - being an extraneous variable that may affect the DV.
What are two methods to minimize research issues (extraneous variables)
- Randomization
- Standardization
What is randomization?
The use of chance methods to control for the effect of bias when designing materials and deciding the order of experimental conditions.
> It reduces the researchers unconscious bias when designing an investigation.
What is the main research issue randomization reduces?
Investigator effects.
What is standardization?
Using exactly the same formalized procedures and instructions for ALL participants in the research study.
This ensures that non standardized changes in procedure do not act as extraneous variables.
What is meant by ‘Experimental design’
Experimental designs refers to ways in which participants are used in experiments. This is how they are arranged / organized in relation to experimental conditions.
What are 3 types of experimental design?
- Independent groups
- Repeated measures
- Matched pairs
What is an independent groups design.
An experimental design in which participants are allocated to different groups, where each group represents one experimental condition (such as control or experimental)
Evaluate Independent group design.
- Order effects
Using independent group design eliminates the problem of order effects (fatigue) - Participant variables
Participants are not the same in terms of participant variables - if a researcher finds a mean difference between groups in the dependent variable, this may be due to participant variables as opposed to the IV. This confounds and reduces the validity of the findings. - Less economical and efficient
More people are needed to produce equivalent data to what would have been produced by a repeated measures design. So, it increases the time and money spent on recruiting participants who would only produce a single result.
What does a repeated measures design involve?
In a repeated design, all participants take part in ALL conditions of the experiment.
Evaluate Repeated measures design.
- Controlled participant variables
As the same participants are used in each condition, participant variables (such as individual differences) are minimized - creating higher validity. - Order effects
The order of conditions can effect participant behavior. This can be in the form of practice effect where the second condition could have better performance as the participants know what to do or fatigue effect where performance deteriorates on the second task. Order effects, therefore are CONFOUNDING variables. - More time-efficient and economical
Fewer participants are needed, so less time and money is spent recruiting them. - Demand characteristics
It is more likely that participants would work out the aim of the study if they had experienced all conditions - creating an extraneous effect.
What is a matched-pairs design?
Where two participants are matched together based on variables that can effect the DV, and then allocated to separate groups. - This then is run like an independent groups design.
Evaluate matched pairs design.
- Order effects and demand characteristics are reduced
The participants only take part in a single condition - Time consuming and expensive
If a pre-test is required to match pairs, this can extend the time taken to carry out the investigation and cost more. - Participant variables not fully eliminated
Even with matched pairs effort to reduce participant variable effects, no participant can ever be matched exactly. Even monozygotic twins have individual differences able to impact the DV. - Counterbalancing not necessary.
As participants only take part in one condition, there is no effort of counterbalancing required to reduce order effects.
What method is used to manage independent groups?
Random allocation
What method is used to manage repeated measures design?
Counter-balancing (ABBA technique).
What does random allocation ensure?
Random allocation ensures that participant variables are addressed through evenly distributing participants across the experimental condition using random techniques such as pulling numbers out of a hat.
What does counterbalancing do in a repeated measures design?
Counterbalancing ensures that order effects are minimized.
Half the participants, for example, take part in condition A then B and the other half take part in condition B then A.
(It doesn’t prevent the problem, it attempts to balance out effects).
What are the 4 types of experiment?
- Laboratory experiment
- Field experiment
- Natural experiment
- Quasi-experiment
What is a laboratory experiment?
An experiment that takes place in a highly controlled environment.
What does a laboratory experiment allow?
It allows the researcher to manipulate the IV to record the effect on the DV, maintaining control of extraneous variables.
What are the strengths of laboratory experiments?
- Control of extraneous and confounding variables
From this, the researcher can ensure any effect on the dependent variable is due to only the manipulation of the independent variable - allowing the establishment of a ‘cause and effect’ relationship. This has HIGH INTERNAL VALIDITY. - High repeatability.
Replication is more possible in lab experiments compared to the others due to high control. This ensures that new extraneous variables aren’t introduced when repeating the experiment. This is vital to check the validity of the findings, showing it was not a one off result.
What are the limitations of laboratory experiments?
- Lacks generalizability
The laboratory environment is most often artificial, not seen in everyday life. This environment may prompt participants to behave unnaturally so their behavior shouldn’t be generalized beyond the research setting - having LOW EXTERNAL VALIDITY. - Demand characteristics
Participants are usually aware that they are being tested in a lab so may behave unnaturally. - Low mundane realism
Lab experiments usually will use artificial stimuli to complement the artificial setting, this means that the tasks the participants are asked to carry out may not represent everyday experience.
What is a field experiment?
An experiment that takes place in a natural setting within which the researcher manipulates the IV and records changes in the DV.
What is the strength of field experiments?
- Higher mundane realism
Compared to lab experiments, the environment of field experiments is more natural - from this, field experiments produce more valid and authentic behavior, especially when done covertly. This gives high external validity.
What are the limitations of field experiments?
- Loss of control over extraneous and confounding variables.
As the setting is more natural, there is lack of control over any extraneous variables that would be minimized in a laboratory setting. This means that cause and effect is more difficult to establish and precise replications are often not possible. - Ethics
Field experiments, when conducted covertly, means that participants may be unaware that they are being studied - hence, not being able to give consent.
This means such research can constitute an invasion of privacy.
What is a natural experiment?
An experiment where the change in the IV is not brought about by the researcher, instead it happens naturally and the researcher records its subsequent effects on the DV they have decided on.
For example, the researcher has no control over the IV of a child being in hospital and is unable to manipulate it.
What and what is not natural in a natural experiment?
- The IV is naturally occurring, unable to be manipulated by the researcher.
- The setting is not necessarily natural - participants may still be lab tested.
- The DV also depends, it may be naturally occurring such as exam results or devised by the experimenter and measured either by field or lab.
What are the strengths of natural experiments
- High external validity
Natural experiments have high external validity due to involving the study of real-world issues as they occur, such as stress levels and disasters. - New research opportunities
Natural experiments give research opportunities that would have not been present before due to practical or ethical purposes, such as the institutionalized Romanian orphan study.
What are the limitations of natural experiments?
- Reduced research opportunity
Naturally occurring events, as studied in natural experiments, may be rare which limits the space for generalizing findings to other similar situations. - Lack of random allocation
Random allocation is only present in independent group designs, meaning the researcher be less sure whether the IV affected the DV. For example, in the Romanian orphan study, the IV was the lateness of when they were adopted. Despite this, there were lots of differences (participant variables) between these two groups as those who were adopted late would be more unsociable and less appealing for adoptive parents.
What is a quasi-experiment?
A quasi-experiment is not exactly an experiment, in this the IV is not determined by anyone - instead, being based on pre-existing differences between individuals such as being young or old. In this, the independent variable cannot be changed.
Evaluate quasi-experiments.
- Highly controlled
Minimizes extraneous variables and enhances repeatability. - No random allocation
Like natural experiments, quasi-experiments do not randomly allocate participants so there may be confounding variables - IV cannot be manipulated.
As the researcher is unable to manipulate the IV, we cannot claim that the IV has caused any observed change.
Distinguish between internal and external validity
Internal validity is about the design of the experiment, what is going on - such as the control.
External validity is about how the findings are able to be generalized from studies to everyday life situations.
What is the difference between a natural and field experiment
- In a field experiment, the researcher is able to manipulate the IV whereas in a natural experiment the researcher has no control over the IV - something else causes this variance.
- In a field experiment, the setting is natural where the researcher goes to the participants environment. For natural experiments, the setting does not necessarily have to be natural, only the IV is definitely naturally occurring.
Distinguish between a quasi-experiment and a natural experiment.
- A natural experiment is where the IV would have occurred regardless of the investigators presence - the IV is naturally occurring
- A quasi-experiment has also has non manipulated IVs, however the IVs are based on pre-existing differences between participants
Define ‘population’
A group of people who are the focus of the researchers interest, from which a smaller sample is drawn
Define ‘sample’
A group of people who take part in a research investigation, drawn from a ‘target population’
Why is sampling done?
For practical and economic reasons, one cannot include every single member of the target population so a smaller group is selected, creating a sample.
What should a sample be and why?
Representative - the sample should be representative of the target population so that findings can be generalized more appropriately.
What are the 5 sampling techniques?
- Random sampling
- Systematic sampling
- Stratified sampling
- Opportunity sampling
- Volunteer sampling
What is a random sample?
Sampling in a way that ensures that all members of the target population have an equal chance of being selected.
Outline the procedure taken when doing a random sample
1 ) Obtain a complete list of all members within the target population
2 ) Assign the names on the list a number
3 ) Select the sample through use of lottery method, such as picking numbers out of a hat or using a randomizer.
Evaluate random sampling
- Potentially unbiased
Random samples are potentially unbiased, this means confounding and extraneous variables would be equally divided between the groups, enhancing internal validity. - Difficult and time consuming.
The method of random sampling can be difficult and time consuming in the sense that a complete list of the target population may be difficult to get, especially concerning the privacy of people. - Does not enhance representativeness
Even using random techniques, there is still a chance for the sample to be unrepresentative. Probability suggests that random sampling should more likely produce a representative sample compared to opportunity sampling, however it is not impossible that random technique can still select people from a similar sub-demographic. - Refusal
Even after the difficulty and effort to gain a list of the target population, participants may still refuse to take part - this makes the sample become more volunteer-like compared to random.
What is systematic sampling
A sampling technique in which every nth member of the target population is selected
(for example, every 5th pupil on the register)
Outline the procedure taken when doing a systematic sample
1 ) Produce a sampling frame into which the list of people in the target population are organized into (usually alphabetical)
2 ) Set up a sampling system, for example picking every 3rd or 6th person.
3 ) Work through the sampling frame until the sample is complete.
Evaluate systematic sampling
- Objective
Systematic sampling is objective as once the system for selection has been established, the researcher has no influence over who is selected - reducing bias. - Time consuming.
The systematic method is time consuming as setting up a frame and system to achieve this may take a long time to gather participants and, in addition, they can also refuse.
What is stratified sampling
A sampling technique in which the composition of the sample reflects the proportions of people in certain subgroups (strata) of the target population or general population.
Outline the procedure taken when doing stratified sampling
1 ) The researcher identifies the different strata that make up the population that will be sampled
2 ) The proportions needed for the sample to be representative are calculated.
3 ) The participants making up each stratum are then RANDOMLY sampled.
Evaluate stratified sampling
- Representative
Stratification produces a representative sample due to the design accurately reflecting the true composition of the population. This means that generalization of findings is possible. - Cannot be fully generalized.
Stratification, however is not perfect. The identified strata is still subject to participant variables so complete representation of the target population is not possible.
What is opportunity sampling?
Opportunity sampling is where researchers select participants based on who is willing and available.
Outline the procedure taken when doing opportunity sampling.
Opportunity sampling could be achieved simply by approaching or asking whoever is around at the time of study, for example on the street.
Evaluate opportunity sampling.
- Convenient
Opportunity sampling is convenient. The methodology is less costly in money and time compared to random sampling due to not requiring a list of target population members and there is no need for calculation and division of these into strata. - Bias
Researcher bias can play a role in opportunity sampling. The researcher has complete control over the selection of participants leading to conscious, or unconscious favor of those who look more approachable and avoidance of those they do not like the look of. - Unrepresentative.
The sample would suffer from being unrepresentative of the target population as it may be drawn from one specific area such as a singular high street - therefore, it cannot be generalized.
What is a volunteer sample?
A sampling technique in which participants select themselves to partake in the sample, also called self-selection.
Outline the procedure taken when doing a volunteer sample
The researcher may place an advertisement in a newspaper or on a notice board. In addition, participants may even raise their hand physically to volunteer when asked by the researcher.
Evaluate volunteer sampling
- Minimal effort.
Collecting a volunteer sample is easy, requiring minimal input from the researcher (therefore, avoiding researcher bias) so it is less time consuming than other forms of sampling. - Volunteer bias
Volunteer sampling is susceptible to volunteer bias. Asking for volunteers as opposed to seeking them out through alternative sampling techniques may attract a specific profile of person. Volunteers may be more likely to ‘please’ the researcher and alter their behavior which then affects how far the findings can be generalized.
What are ethical issues?
Conflicts existing between the rights of participants in research studies and the goals of the researcher to produce authentic and worthwhile data.
What are the four major ethical issues in research?
- Informed consent
- Deception
- Protection from harm
- Privacy and confidentiality.
What is meant by informed consent?
Making participants aware of the aims of research, procedures, their rights in relation to withdrawal and what the data would be used for.
Following this, the participant should make an informed judgement whether to take part in the experiment or not without coercion.
What is meant by deception?
Deception is deliberately misleading or withholding information from participants at any stage of the investigation.
What is meant by protection from harm?
An ethical issue that states that as a result of their involvement, participants should not be subject to any more risk than they would be in their daily lives - being protected from physical and psychological harm.
What is meant by privacy and confidentiality?
Privacy and confidentiality refers to the right to control information about ourselves (privacy) and to have any personal data protected (confidentiality).
What is the BPS code of conduct?
A quasi-legal document produced by the British Psychological Society that instructs UK psychologists on acceptable and non-acceptable behavior when dealing with participants.
What does the BPS code of conduct do?
The BPS code of conduct acts as guidelines, meaning a researcher has a professional duty to abide by these when conducting research.
The guidelines attempt to ensure participants are treated with respect and consideration in regards to the ethical issues present.
How does the BPS decide whether proposals are ethically acceptable?
Guidelines are implemented by ethics committees who use a COST-BENEFIT ANALYSIS to determine whether research proposals are ethically acceptable.
How would you deal with informed consent?
Participants should be issued with a consent letter/form detailing all relevant information that could affect their decision to participate. For investigations involving children under 16, parental consent is required?
