Research Methods Flashcards
What are confounding variables?
Confounding variables change systematically with the IV, so confound the findings of an experiment.
What are extraneous variables?
- EV’s are additional, unwanted variables that could potentially interfere with the IV
- Many EV’s are straightforward to control, such as participant age and lighting in the lab
- These are described as ‘nuisance variables’ that do not vary systematically with the IV
- They do not confound the findings of the study, but may make it harder to detect a result
What is mundane realism?
The term mundane realism refers to how an experiment mirrors the real world. If an experimental task lacks mundane realism, the results of the study may not be very useful in terms of understanding behaviour in the real world.
What is generalisability?
The ability of your results to be able to be generalised to the full population. The materials and environment of the study effects it’s generalisability.
What is internal validity?
Internal validity concerns what goes on inside a study.
- Is the experiment testing what it is meant to be testing
- Did the IV produce a change in the DV or did something else affect it
- Were there any confounding variables
- Did the study have mundane realism
What is external validity?
External validity is affected by low internal validity, as you cannot generalise the results of a study that was low in internal validity because the results have no real meaning for the behaviour in question.
Ecological validity
Population validity
Historical / temporal validity (over time)
What is a directional hypothesis?
States the expected direction of the results, i.e. you are stating that people who sleep well do better on class tests
What is a non-directional hypothesis?
States there is a difference between two conditions, but does not state the direction of the difference, i.e. people who have plentiful sleep have different marks on a class test
What are pilot studies?
- A small scale trial run of the investigation which takes place before the real one is conducted
- The aim is to check the procedure, materials, et cetera to allow the researcher to iron out any potential problems and make any modifications that may be necessary
- They are usually run with a handful of participants and can be used to test out any research method from experiments to questionnaires to observations
- They are important because the opportunity to identify and modify the investigation will save any time and money that could possibly be wasted during the real thing
Outline independent groups
Different groups of people performing the conditions
+ Reduced demand characteristics
+ No order effects
- Participant and/or situational variables
- More participants needed
Outline repeated measures
One group performs all the conditions
+ No participant variables
+ Less participants needed
- Demand characteristics likely
- Could be order effects (practice, fatigue)
Counterbalancing - ABBA or AB BA
Outline matched pairs
Participants are paired up on a certain quality, such as age or gender
+ Reduces participant variables
+ Reduces demand characteristics
+ No order effects
- More participants needed
- Matching participants can be time consuming and difficult
Outline lab experiments
Lab experiments are experiments conducted in a special environment where variables can be carefully controlled.
+ Controlled so high internal validity
+ Easily replicated
- Low mundane realism
- Low ecological validity
- More likely to have demand characteristics, participant effects, and investigator effects
Outline field experiments
Field experiments are conducted in a real life setting. The IV is still deliberately manipulated by the researcher.
+ Often unaware they are being studied so less demand characteristics
+ Higher mundane realism and ecological validity
- More difficult to control ev’s
- Ethical issues
- Less easy to replicate
Outline natural experiments
A natural experiment is conducted when it is not possible, for ethical or practical reasons, to deliberately manipulate the IV. The IV is a naturally occurring event.
+ Allows research where IV can’t be manipulated due to ethical or practical reasons
+ Increased mundane realism and ecological validity
- Can’t demonstrate causal relationships
- Random allocation not possible so may be variables that can’t be controlled
- Probably unable to replicate
Outline quasi experiments
IV is based on existing personal differences (age, gender)
+ Allows comparison between types of people
+ Less experimenter bias
- Demand characteristics so reduced internal validity
- Decreased mundane realism
- Increased chance of participant variables
What are demand characteristics?
- Participant reactivity is a significant extraneous variable in experimental research and is very difficult to control
- In a research situation, participants will try and work out what is going on, using certain clues
- These clues are the demand characteristics of the experimental situation and may help a participant ‘second guess’ the experimenters intentions as well as the aims of the study
- Participants may look for clues to tell them how they should behave in the experimental situation
- They may act in a way they think is expected and over-perform to please the experimenter, which is the ‘please-you effect’
- They may deliberately under-perform to sabotage the results of the study, which is the ‘screw-you effect’
- Either way, participant behaviour is no longer natural, and thus and extraneous variable that may affect the DV
What are investigator effects?
- Investigator effects are any cues (other than the IV) from an investigator that encourage certain behaviours in the participant, and which might lead to the fulfilment of the investigator’s expectations
- Such cues act as extraneous or confounding variables
Indirect investigator effects:
The ‘investigator experimental design effect’ -> the investigator may operationalise the measurement of variables in such a way that the desired result is more likely, or may limit the duration of the study for the same reason.
The ‘investigator loose procedure effect’ -> this refers to situations where an investigator may not clearly specify the standardised instructions and/or procedures, which leaves room for the results to be influenced by the experimenter.
How do you reduce experimenter bias and demand characteristics?
Single blind - withhold aim from participants
Double blind - neither the participant or researcher is aware of the aim
Control group - neutral group to formulate comparisons or set a baseline
What are participant variables?
Any characteristic of individual participants, such as personality or intelligence.
What are situational variables?
Features of a research situation that may influence participants’ behaviour, such as temperature or time of day.
What are the five types of sampling?
Opportunity
Random
Stratified
Systematic
Volunteer
Outline opportunity sampling
People who are most convenient / available.
+ Easy, not time consuming
- Biased because sample drawn from a small part of the population
Outline random sampling
Sample obtained using random techniques such as the lottery method or the random number table.
+ Unbiased
- May be time consuming, need a list of the population
Outline stratified sampling
Subgroups within a population are identified, for example gender or age, and participants are obtained from each of the strata in proportion to their occurrence in the population. Selection from the strata is done randomly.
+ Representative
- Time consuming
Outline systematic sampling
Using a predetermined system to select participants, such as selecting every nth person from a phone book.
+ Unbiased
- Time consuming
Outline volunteer sampling
+ May be more representative and less biased
- Participants more likely to be highly motivated to be helpful or want the reward
- Volunteer bias
What are the six ethics guidelines?
Informed consent
Deception
Confidentiality and privacy
Debrief
Right to withdraw
Protection from (psychological) harm
What are the strategies of dealing with ethical issues?
A cost-benefit analysis -> the costs of doing the research is judged against the benefits. The costs and benefits can be judged from the point of view of participants, society as a whole, or in terms of the group to which an individual belongs
Ethics committees -> these committees must approve any study beforehand, and the committee looks at all possible ethical issues raised in any research proposal and how the researcher suggests those issues will be dealt with
Punishment -> if a psychologist behaves in an unethical manner, the BPS reviews the research and may decide to bar the person from practicing as a psychologist.
Outline naturalistic observations
In a naturalistic observation, behaviour is studied in a natural situation where everything has been left as it is normally.
+ Gives a realistic picture of spontaneous behaviour
- Little control over variables
Outline controlled observations
In a controlled observation, some variables in the environment are regulated by the researcher, reducing the ‘naturalness’ of the environment, and the ‘naturalness’ of the behaviour being studied.
+ An observer can focus on particular aspects of behaviour
- Control of variables means environment and behaviour is less natural
Outline covert observations
People are observed without their knowledge.
+ Less demand characteristics - increased validity
+ Natural behaviour
- Ethical issues
Outline overt observations
Participants are aware their behaviour is being studied.
- Demand characteristics
- Unnatural behaviour
Outline participant observation
Observations are made by someone who is also participating in the activity being observed.
+ Provides special insights into behaviour that might not otherwise be gained
- Likely to be overt and thus have issues of participant awareness
Outline non-participant observation
The observer is separate from the people being observed.
+ Observers are likely to be more objective
- Likely to be covert, and thus ethical issues may arise
What are unstructured observations?
All relevant behaviour is recorded with no system. An issue is there may be too much too record. Furthermore, the behaviours recorded will often be those which are the most visible or eye-catching to the observer, but these may not be the most important or relevant behaviours. Researchers may use this approach where research has not been conducted before, as a kind of pilot study to see what behaviours might be recorded.
Outline structured observations
The researcher has a system used to record behaviour.
Behavioural categories:
Behaviour being studied is broken up into a set of components. Behavioural categories should:
- Be objective
- Cover all possible component behaviours and avoid a ‘waste basket’ category
- Be mutually exclusive
Sampling procedures:
Event sampling - counting the number of times a certain behaviour occurs
Time sampling - recording behaviours in a given tome frame, for example, noting what the participant is doing every 30 seconds
What is operationalisation?
Turning the DV into a measurable format
Evaluate self-report techniques
+ Allows access to what people think and feel, and to their experiences and attitudes
- People may not supply truthful answers due to social desirability bias
- If people do not know the answer to a question, they often make it up, meaning answers may lack validity
- The sample of people used in any study using self-report may lack representativeness and thus the data may lack generalisability
Outline questionnaires
Questionnaires are a self report technique. They allow researchers to discover what people think and feel, in contrast to observations, where researchers have to ‘guess’ what people think and feel on the basis of how they behave.
+ Once designed and tested, questionnaires can be distributed to large numbers of people relatively cheaply and quickly, enabling a researcher to collect data from a large sample of people
+ Respondents may be more willing to give personal information in a questionnaire than an interview, where they may feel self-conscious and more cautious
- Questionnaires are only filled in by people who can read and write and have the time to fill them in, meaning the sample is biased
Outline structured interviews
A structured interview has pre-determined questions, so is essentially a questionnaire that is delivered face to face. It is a self-report technique.
+ The main strength of a structured interview is that it can be easily repeated because the questions are standardised, meaning answers from different people can be compared
+ This also means that they are easier to analyse there an unstructured interview because answers are more predictable
- Comparability may be a problem in a structured interview, if the same interviewer behaves differently on different occasions or different interviewers behave differently
- A limitation of both structured and unstructured interviews is that the interviewers expectations may influence the answers the interviewee gives - interviewer bias
Outline unstructured interviews
In an unstructured interview new questions are developed during the course of the interview. It is a self-report technique.
+ More detailed information can generally be obtained from each respondent than in a structured interview because the interviewer tailors further questions to the specific response
- Unstructured interviews require interviewers with more skill than a structured interview because the interviewer has to develop new questions on the spot
- Such in-depth questions may be more likely to lack objectivity than predetermined ones, because of their instantaneous nature, with no time for the interviewer to reflect on what to say
How are questionnaires constructed?
There are three guiding principles when writing questions:
1. Clarity -> there should be no ambiguity in questions
2. Bias -> any bias in a question may lead the respondent to be more likely to give a particular answer, with the biggest issue being social desirability bias
3. Analysis -> questions need to be written so the answers are easy to analyse, so closed questions would be better to analyse
