Research methods Flashcards
Define experiment
Tests undertaken under controlled conditions in order to investigate the effect of the IV on the DV
Define lab experiment
Experiments that occur in a highly controlled setting where the experimenter directly manipulates the IV
Define field experiment
Experiments that occur in a natural environment of the people whose behaviour is being studied
Define quasi experiment
Where the IV is naturally occurring as opposed to be created and controlled by the experimenter
List the strengths of lab experiments
Control in a lab experiment will produce scientific research, which ensures that the variable which we are manipulating is the only thing affecting behaviour (ie no extraneous variables)
List the weaknesses of lab experiment
Lower EV as they are artificial settings which do not reflect real life, so therefore the behaviour seen may also be artificial
What are the strengths of a field experiment?
More realistic setting for a study and can have higher EV
What are the weaknesses of a field experiment?
Lack of control can mean it is difficult to assume the IV was actually influencing behaviour and that it wasn’t another extraneous variable
What are the strengths of quasi experiments?
Allows us to study the effects of the variables psychologists can’t manipulate or change behaviour
What are the weaknesses of quasi experiments?
There is no control over the participants, in terms of social setting, how they were brought up, lifestyle etc., and these may be confounding variables which can influence behaviour
Define repeated measures
This involves using the same P’s in each condition
Define independent measures
This involves using different people in each condition
Define matched groups
This involves using different people in each condition but an attempt is made to make the P’s as similar as possible on certain key characteristics
What are the advantages of repeated measures?
Reduces individual differences as P is being compared to themselves and uses fewer P’s so maybe more cost and time effective
What are the disadvantages of repeated measures?
Must be counterbalanced due to order effects and P’s may work out the IV and display DC’s
What are advantages of independent measures?
Unaffected by order effects as each P is only tested once, less likely to be affected by DC’s and less time consuming than matched groups
What are the disadvantages of independent measures?
Individual differences may skew the results and findings, large sample needed to be sure the DV is caused by IV and not by individual differences
What are the advantages of matched groups?
Unaffected by order effects or DC’s, controls P variables better because P’s are matched on variables important to the study
What are the disadvantages of matched groups?
Very time consuming to match P’s, impossible to match P’s on enough variables to be sure that there are no possible extraneous variables or individual differences
Define self selecting sampling method
This is when people volunteer to take part in the study. Often adverts, posters or leaflets will be distributed which contain details about the research
Define opportunity sampling method
A sample of P’s produced by selecting those who are most readily available at a given time and place selected by the researcher
Define random sampling method
A technique in which each member of the target population has an equal chance of being selected (e.g. using a random name generator)
Define snowball sampling
When P’s are asked to contact their friends and family and ask them to also take part in the research
What are the strengths of self selecting?
Can target certain groups you want
Relatively easy and P’s are likely to turn up
What are the weaknesses of self selecting?
Biased based on who volunteers (more confident people?) and possible time and cost of the advertising process
What are the strengths of opportunity sampling?
Quick to gather P’s and very easy as no advertising or selection process required
What are weaknesses of opportunity sampling?
Biased based on where you go and who is available at the time
May be less ethical as P’s may feel more obliged to take part
What are strengths of random sampling?
Should be representative of the target population and all have chance of being selected
What are weaknesses of random sampling?
Can be difficult to ensure all names from target pop are included
Can still be biased on who (from those selected) are willing to take part
What are strengths of snowball sampling?
Easy as only requires finding a few P’s before they recruit the rest
Relatively cheap
Good when after ‘rare’ groups
What are weaknesses of snowball sampling?
Biased as all P’s know each other and are likely to have many similarities
May not get enough P’s
What is quantitative data?
Quantities of things (e.g. raw scores, numbers and percentages)
What is qualitative data?
Qualities of things (e.g. descriptions, words, meanings, pictures)
What are the strengths of quantitative data?
Easy comparisons
Summarises easily into percentages
Reliable and replicable
What are the weaknesses of quantitative data?
Not reflective of everyday life
EV us lacking
No insight into why results occurred
What are the strengths of qualitative data?
Richness and details to the findings
Higher levels of validity
What are the weaknesses of qualitative data?
Harder to summarise
Difficult to make comparisons
Define ‘measures of central tendency’
These all aim to express the ‘average’ or ‘typical’ score within a data set
What are the 3 measures of central tendency?
Mean, median and mode
What are the advantages and disadvantages of the mean?
+: Accounts for all of the results obtained
-: Allows for extreme results and can be skewed or distorted
What are the advantages and disadvantages of the median?
+: Eliminates extreme scores
-: Could be a result that doesn’t exist
What are the advantages and disadvantages of the mode?
+: Represents the majority of results and works for non-numerical data
-: Doesn’t consider all values
Define ‘measures of dispersion’
They indicate how far the results are spread out around the ‘typical’ score
What are the 3 measures of dispersion?
Range
Variance
Standard deviation
How is the variance calculated?
Subtract each score from the mean score (this is ‘d’)
Square each ‘d’ score
Find the mean of the ‘d’ squared’s
What are the advantages and disadvantages of the range?
+: Shows us the range of values in the data
-: Can be skewed by extreme values and doesn’t account for the distribution of data
What are the advantages and disadvantages of the variance?
+: A more precise calculation as all points are included and less likely to be skewed by extreme values
-: It is a square figure so is not represented in the same unit as the original data
What are the advantages and disadvantages of the standard deviation?
+: Precise calculation as all data is included, less likely to be skewed and score is represented in the same unit as original data
-: Time consuming and more difficult
Define ‘observation’
Where the researcher observes the behaviour of the P’s and records it
What are the general strengths of observations?
Less intrusive than experiments
Low DC’s and investigator effects if P’s are unaware they are being observed
High EV
Can be used when other methods may not be appropriate
What are the general weaknesses of observations?
Lack of control over the situation/environment can lead to an inability to establish cause and effect
Difficult to replicate
If P’s are aware they may change behaviour
Problems with reliability due to possible observer bias
What is a covert observation?
Where P’s are unaware that they are being observed
What are the strengths of covert observations?
High EV as usually occurs in real life setting
Low DC’s as P’s are unaware they are being observed
What are the weaknesses of covert observations?
Limited control over extraneous variables
Difficult to generalise
Possible observer bias/misinterpretation
Possible ethical issues
What is an overt observation?
Where P’s are aware that they are being observed
What are the strengths of an overt observation?
Ethical as P’s are aware of the observations and the presence of observer becomes habituated after a while
What are the weaknesses of an overt observation?
Behaviour may change due to the presence of the observer and possible DC’s
What is a naturalistic observation?
Where research is carried out in the P’s natural environment
What are the strengths of a naturalistic observation?
Study is realistic and therefore there is higher EV
Observed behaviour is more natural, and if covert, less DC’s
Can be used when it may be difficult to use other methods (e.g. questionnaires)
What are the weaknesses of naturalistic observations?
Can be difficult to control extraneous variables Difficult to record data - esp. if covert Ethical concerns (if covert)
What is a controlled observation?
Where research is carried out in conditions arranged by researcher, such as in a laboratory setting
What are the strengths of a controlled observation?
High control allows researcher to focus on the specific behaviour of interest
Ethical as P’s will have consented
Replication is possible if there is a standardised procedure
What are the weaknesses of a controlled observation?
Presence of researcher may influence behaviour and low EV as an artificial environment (so cannot be generalised)
What is a participant observation?
Where the researcher interacts with the people they are observing as an ‘inside’ observer and the group is unaware of the researchers real purpose
What are the strengths of a participant observation?
High EV
Participation allows for deeper understanding
Can observe behaviour which may usually be hidden from outsiders
What are the weaknesses of a participant observation?
Possibly unethical
May be difficult to record data
Presence of observer may influence behaviour
What is a non-participant observation?
Researcher observes from a distance
What are the strengths of a non-participant observation?
Ethical (if overt or in a public place) and observer may remain more objective as they are not involved
What are the weaknesses of a non-participant observation?
May not get the real meaning of behaviour observed
May be problems with DC’s (if overt)
May be practical difficulties in seeing and recording behaviour from a distance
What is time sampling?
Where behaviour is recorded at a set interval (e.g. every 10 minutes)
What are strengths of time sampling?
Observation is more structured as no issues with trying to record all behaviour and therefore better validity
Fewer observers needed so cheaper
Can be used in situations where event sampling wouldn’t work as well
What are the weaknesses of time sampling?
Need to have an accurate time keeping device to know when each interval starts and finishes
May be unrepresentative as only showing behaviour in a certain time period
What is event sampling?
When you record each time the target behaviour occurs throughout the duration of the observation
What are the strengths of event sampling?
More accurate representation of the range of behaviours that occur as every occurrence of all the behaviours are recorder
No need for precision timing
Useful when you want to see all behaviours present
What are the weaknesses of event sampling?
May be difficult to record all of the behaviours accurately (issues with inter-rater reliability)
Will need well trained observers to ensure accurate recording (time consuming and expensive)
Not easy to do when looking at a large number of behaviour
What is an open question?
When P’s are asked to respond to a set question but leaves a blank underneath the Q so P’s can answer in whatever style they choose and produce qualitative data
What are the strengths of open questions?
Qualitative data
Provides insight into people’s feelings and opinions (richness and detail)
Doesn’t force P’s to give a particular response
What are the weaknesses of open questions?
Difficult to code
Difficult to establish reliability of qualitative results
Time consuming
What are closed questions?
When P’s are given answers and they have to choose the most appropriate answer for them
What are the strengths of closed questions?
Easily turned into quantitative data which can be compared and analysed
Easy to test reliability of data
What are the weaknesses of closed questions?
Difficult to know the reasons behind answers
Response set bias
Can lack EV
What are rating scales?
A set of values or categories, ranging from one extreme to its opposite, which allow feelings and behaviours to be rated
What are strengths of rating scales?
More detailed than yes/no
Comparable quantitative data
Easily repeated
What are the weaknesses of rating scales?
Tendency of choosing the middle score
Doesn’t give insight into why P chose that option
Restricted response set
What is a likert scale?
Where people are given a range of answers from which they select the one which represents to the extent which they like/dislike or agree/diagree with something
What is a semantic differential?
A scale which is used to put something on a scale between two descriptive words (e.g. strong and weak)
What are the 3 types of interview?
Structured, semi-structured and unstructured
Define structured interview
The interviewer asks the same questions to each P in the same order. Closed questions are often used, with interviewee picking the answer closest to their own view
What are the strengths of a structured interview?
Using same Q’s means it’s standardised and replicable and responses can be easily compared
What are the weaknesses of a structured interview?
Additional q’s can’t be asked which means further clarification may not be found and interesting points not followed up on
What is a semi-structured interview?
The interviewer will have a set of q’s to ask and is expected to ask them all but the phrasing and timing is up to the interviewer and the questions may have more open-ended responses
What are the strengths of a semi-structured interview?
Allows interviewer to use additional questions to seek clarification or explore comments made during a set question
What are the weaknesses of a semi-structured interview?
Interview still likely to be constrained around the set questions and additional Q’s may vary from P to P so not comparable
What is an unstructured interview?
The researcher has topics to discuss but these don’t have to be in the same order. It is more like a conversation, with further questions being developed in response to the interviewee’s answers
What are the strengths of an unstructured interview?
Allows for info to be gathered which might not be revealed from pre-determined Q’s
Freedom to asks Q’s on the spot
What are the weaknesses of unstructured interviews?
Different responses from different P’s as they may be asked very different q’s
What are the strengths of questionnaires over interviews?
Easily done, quick, comparable answers, quantitative data, replicable
What are the strengths of interviews over questionnaires?
Can get a rapport and can assess body language
What are the weaknesses of questionnaires?
Response rate, may lie (easy to do when not face to face), standard response bias
What are the weaknesses of interviews?
Time consuming, limited sample and less confidential
How/why would leading questions reduce validity and how could this be overcome?
P’s don’t express their genuine views but instead ones they feel led towards, therefore, avoid leading questions
How/why would social desirability bias reduce validity and how could this be overcome?
P’s don’t express real views but instead views which they think make them look good, therefore, don’t ask P’s to record their names
How/why would forced-choice questions reduce validity and how could this be overcome?
P’s don’t express real views but instead the views closest to their real views so therefore should have an ‘other’ or ‘please specify’ option
How/why would demand characteristics reduce validity and how could this be overcome?
P’s don’t express real views but instead express views which they think the researcher is looking for, so therefore use ‘smokescreen’ q’s to disguise the aim of the study
How/why would standard response set (choosing same answer each time) reduce validity and how could this be overcome?
P’s don’t express real views and just give answers without really thinking so therefore, should reverse statements so strong agree doesn’t always express a positive attitude
How/why would P’s recording their names reduce validity and how could this be overcome?
P’s don’t express real views as they feel they can be tracked down so tone down their views, so therefore, should guarantee P’s anonymity
How/why would researcher bias reduce validity and how could this be overcome?
The way that data is analysed might not reflect P’s views and instead what the researcher wanted P’s to say so therefore, get an independent researcher who is ‘blind’ to the aim of the study
Does a large number of questions asking about the same thing improve or reduce reliability and how?
Improves it - this is a good way of testing if something is really true of P (as opposed to being a ‘fluke result’)
If a questionnaire is repeated with several groups and similar results are found, does this improve or reduce reliability and how?
Improves it - this helps to establish high external reliability and consistent findings are being found
If some P’s complete questionnaire on their own and others complete with others present, does this improve or reduce reliability and how?
Reduces it - there is no consistency between P’s in terms of their experience of how they completed the questionnaire
If there is a low response rate, does this improve or reduce reliability and how?
Reduces it - not a large enough sample to establish consistent effect
If there is an unstructured interview meaning different P’s are asked different questions, does this improve or reduce reliability and how?
Reduces it - there is no consistency between P’s in terms of their experience (so P’s results cannot be easily compared)
If there are ambiguous questions, does this improve or reduce reliability and how?
Reduces it - there will be inconsistency between P’s with regard to what they think they have been asked and therefore will respond to different Q’s based on their perceptions
If the researcher compares responses from odd q’s to even q’s, does this improve or reduce reliability and how?
Improves it - there is high split-half reliability which means the questionnaire/interview has internal reliability
What are the strengths of correlations?
Can tell us something new
Useful technique if it may not be practical or ethical to manipulate variables
Can act as a good starting point for further research
May be the only way to investigate variables we cannot control
What are the weaknesses of correlations?
They do not show cause and effect and the stats tests will not always pick up on a relationship between two co-variables (e.g. if parabola was produced)
