Research Methods Flashcards
What are the advantages of laboratory experiments?
CONTROL - minimising effects of confounding variables
REPLICATION -
CAUSAL RELATIONSHIPS - possible to establish if one thing causes another
What are the disadvantages of laboratory experiments?
ARTIFICIAL - they may lack ecological validity
DEMAND CHARACTERISTICS -
ETHICS - deception is often used
What are the advantages of doing Field experiments?
CAUSAL RELATIONSHIPS - you can still establish these by manipulating the key variable and measuring its effect, although its more difficult to do
ECOLOGICAL VALIDITY
DEMAND CHARACTERISTICS
What are the disadvantages of Field Experiments?
LESS CONTROL - confounding variables are more likely
ETHICS
What are the advantages of Natural experiments?
ETHICAL
What are the disadvantages of Natural experiments?
PARTICIPANT ALLOCATION - you have no control so confounding variables will creep in
RARE EVENTS - some groups of interest hare hard to find
ETHICS - has deception been used? Also has confidentially been compromised?
What are the advantages of Naturalistic Observation?
ECOLOGICAL VALIDITY - participant is unaware of being observed
THEORY DEVELOPMENT - can be useful method of developing ideas which can be funnelled into more controlled conditions later
What are the disadvantages of Naturalistic Observations?
EXTRANEOUS VARIABLES - cant control variables that may have affected behaviour
OBSERVER BIAS
ETHICS - debriefing is difficult. Respecting privacy may be difficult. Getting informed consent may be difficult
What is the advantages of Correlational Research?
CAUSAL RELATIONSHIPS - these can be ruled out if no correlation exists.
ETHICS - can study variables that it would be unethical to manipulate
What are the disadvantages of Correlational Research?
CAUSAL RElATIONShIPS - these can not be assumed from correlation
ETHICS - sometimes causality is inferred from correlation
What are the advantages of Questionnaires?
PRACTICAL - can collect a lot of information quickly and cheaply
What are the disadvantages of Questionnaires?
BAD QUESTIONS - leading or unclear questions may be used
BIASED SAMPLES - the people who will respond to questionnaires may be unrepresentative
SELF REPORT - results may be unreliable because of social desirability bias
ETHICS - confidentiality can be a problem, especially around sensitive issues
What are the advantages of interviews?
RICH DATA - you can get detailed information
PILOT STUDIES - interview are a great way to get information before a study
What are the disadvantages of interviews?
SELF REPORT - can be unreliable and effected by social desirability bias
IMPRACTICAL - time consuming and requires skilled interviewers
ETHICS - confidentiality can be a problem especially around sensitive issues
What are the two types of interviews?
Structured interviews - set list of questions
Unstructured interviews - set of discussion topics only
What are the advantages of Case Studies?
RICH DATA
UNIQUE CASES can challenge existing ideas and suggest ideas for further research
What are the disadvantages of using Case Studies?
CAUSAL RELATIONSHIPS - little control over variables
GENERALISATION - results from single cases are difficult to generalise
ETHICS - informed consent can be difficult to obtain if the subject has a disorder
What are the main topics that you should consider when asked about advantages and disadvantages of different research methods?
Ethics Reliability Replicability Controlling variables Sampling Generalising results Identifying causal relationships
What is a research aim?
This a statement about what the study intends to investigate
What is the research Hypothesis?
This is the statement of what is actually being tested. It is proposed at the beginning of a study
Name the two kinds of Hypotheses
Null hypothesis
Experimental (alternative) hypothesis - these can be…..
Non-directional hypothesis
Directional hypothesis
What is the Null Hypothesis?
This is what you are going to assume is true during the study.
Very often the Null Hypothesis is a reduction that there will be no relationship between key variables
What is the Experimental or Alternative Hypothesis?
If the study forces you to reject the Null hypothesis then you accept your Alternative Hypothesis.
So if your Null Hypothesis is that the two variables aren’t linked then the Alternative Hypothesis is that they ARE linked. OR you can be more specific and fine HOW they are linked using Directional or Undirectional Hypotheses instead
When is a Directional Hypothesis used?
These are often used when a previous study indicates which way a the results will go
When is a Non-directional Hypothesis used?
This would predict a difference but not specify which way. It is often used if there is not a lot of previous search in the area or when the searchers are unsure
What are the three types of variables that may be in an experiment?
Dependant variables
Independent variables
Extraneous (or Confounding) variables
What are Independent variables?
These are variables that are directly manipulated by the researcher.
What are Dependant variables?
The DV is the variable you think will be affected by the change in the IV. So the DV is dependent upon the IV.
What are Extraneous Variables?
Extraneous variables are other variables which might effect the DV
If they DO effect the DV then they are called Confounding variables
What is Operationalisation?
Operationalisation is showing how the variables will be measured.
Some variables are easy to operationalise ( height)
Other variables are harder (a mother’s love)
Name the three most common Research Designs
Independent Research Design
Repeated Measures Design
Matched Pairs Design
What is an Independent Groups Design?
There are different participants in each group
What is the advantage of an Independent Groups Design?
NO ORDER EFFECTS - no one can learn through practice or suffer fatigue effects
What are the disadvantages of Independent Groups Design?
PARTICIPANT VARiABLES - difference between people in each group might affect the result
NUMBER OF PARTICIPANTS - you need twice as many participants as for the Independent Variables Design
What is the Repeated Measures Design?
All the participants do all conditions of the test
What are the advantage of Repeated Measures Design?
PARTICIPANTS VARIABLES - the same people do all parts so no individual differences to worry about
NUMBER OF PARTICIPANTS - you need fewer participants that for the Other two research designs
What is the disadvantage of Repeated Measures Design?
ORDER EFFECTS - because the same participants taking part in all conditions then students could become practiced or suffer fatigue effects
What is Matched Pairs Design?
There are different participants for each condition but they are matched for important variables
What are the advantages of the Matched Pairs Design?
NO ORDER EFFECTS - there are different people in each condition
PARTICIPANT VARIABLES - important differences a minimised by matching
What are the disadvantages of the Matched Pairs Design?
NUMBER OF PARtICIPANTS - need twice as many than for the Repeated Measures Design
PRACTICALITIES - it is time consuming finding participants that match
Why would you choose to run a Pilot Study?
To try out your research design and troubleshoot any problems
How can you control variables so their unwanted effects are minimised?
Counterbalancing Random allocation Keeping Extraneous variables constant Eliminating Extraneous variables Using Standardised Instructions
Why would a Repeated Measures Design use Counterbalancing of variables?
Mixing up the order of tasks in an experiment can solve Order Effects in a Repeated Measures Design
What good does Random Allocation do in a Independant Measures Design?
Random allocation means that everyone has an equal chance of being selected to do either condition - should eliminate bias
What factors would researchers consider to decide if a test had INTERNAL RELIABILITY?
INTERNAL RELIABILITY is when a test gives consistent results within itself. The SPLIT-HALF technique assesses this. A questionnaire is split in half - if all participants score similarly on both halves then the experiment has Internal Reliability
What factors would a researcher consider in order to assess whether an experiment had EXTERNAL RELIABILITY ?
If the measure is stable over time or between people then the experiment has External Reliability.
This can be assessed by measuring
TeST-RETEST Reliability (does the same person always give similar results on the test)
INTER-RATER Reliability (do different assessors give the same conclusions?)
What factors do researchers have to consider when assessing the INTERNAL VALIDITY of an experiment?
If an experiment shows that the results were caused by the manipulation of the variables rather than the effect of something else then it has Internal Validity
What factors do researchers have to consider when assessing the EXTERNAL VALIDITY of an experiment?
If the findings can be generalised beyond the experimental setting than the experiment has External Validity
What is Participant Observation?
This is when the researcher participates in the activity unr study in an overt way.
What are the advantages of a Participant Study?
The researcher develops a relationship with the group under study
What is the disadvantage of a Participant Study?
The researcher can lose objectivity
Participants can react differently if they know that a researcher is amongst them
What is a Non-Participant Study?
This is when the researcher just observes covertly without getting involved
What are the advantages of a Non-participant study?
The researcher can remain objective throughout
What are the disadvantages of a Non-Participant Study?
The researcher loses some sense of the group dynamics by not being involved
What are Structured Observations?
This is when the behaviour categories to be used are identified beforehand - researchers have to tick boxes
What is the advantage of Structured Observations?
It’s easier to gather relevant data because you already know what you are looking for
What are the disadvantages of Structured Observations?
Interesting behaviours could go unreported because they haven’t been per-defined as important
What 5 things do you need to consider when designing a Naturalistic Observation?
Recording data (written, video, audio?)
Categorising behaviour - defining and operationalising
Rating behaviour - deciding on degree or coding behaviour - gives quantitative data
Sampling behaviour - how often and how long to observe for I.e.
Event sampling (only focussing on behaviour you are interested in) OR
Time-interval sampling (observing during set time intervals)
Inter-Observer Reliability
What is the advantage and disadvantage of Event Sampling?
Advantage - researches know exactly what behaviour to look for
Disadvantage - potentially interesting behaviours could be ignored
What is the advantage and disadvantage of Time-Interval Sampling?
Advantage - very convenient for researchers
Disadvantage - could miss interesting behaviour if it doesn’t happen in the specified time interval
How would you ensure observer reliability?
Inter-Observer Reliability is assured by comparing data from two or more observers to make sure that they are giving the same scores
Need to train the observers and ensure inter-observer reliability before actual observation
What are the 5 things to consider when designing a questionnaire or an interview?
- Type of data you want to collect - qualitative or quantitative and thus type of questions asked (open or closed)
- Ambiguity - questions must be specific not vague
- Double-barrelled questions - avoid questions that ask two things
- Leading questions - avoid
- Complexity - use clear English and avoid jargon
What are open questions?
Questions where the participant can reply in any way. This gives detailed answers but hard to quantify and to analyse
What are closed questions
These limit the answers that can be given. They give quanta give data, easier to analyse, but not so much qualitative data
What special things do you also have to consider when design g an interview?
- How structured the interview will be - informal with few set questions. This gives detailed qualitative data but not quantitative or easy to analyse. Or structured with set questions and closed answers yielding more quantitative data
- Using a question checklist
- The behaviour or appearance of the interviewer - this could influence how the participants react
What 3 ways are there of selecting a sample of participants?
Random Sampling
Opportunity Sampling
Volunteer Sampling
What is Random Sampling?
Where every member of the target group has an equal chance of being selected for the sample.
What are the advantages and disadvantages of Random Sampling?
Advantages. It is fair and the sample is likely to be representative
Disadvantages. A representative sample isn’t guaranteed.
IN PRACTICE A RANDOM SAMPLE IS RARELY USED
What is Opportunity Sampling?
When researcher samples whoever is available and willing to be studied (such as students in a university)
What are the advantages and disadvantages of Opportunity Sampling?
Advantages. Quick and practical
Disadvantage. Unlikely to be representative of the target population so we can’t generalise the results
OPPORTUNITY SAMPLING IS OFTEN USED
What is Volunteer Sampling?
When people volunteer to be in a study - suitable candidates are then chosen from these
What are the advantages and disadvantages of Volunteer sampling?
Advantages. If advertised widely then a large number of participants may volunteer. This would allow in-depth analysis and more accurate statistical results
Disadvantages. May still not be representative of the target population. Also, people who volunteer may be more cooperative than others.
If people know that they are being studied they may show 3 effects. Name them.
The Hawthorne Effect
Demand Characteristics
Social Desirability Bias
What is the Hawthorne Effect?
If people are interested in something and receiving positive attention they may try harder. The results will then be artificially high and inaccurate.
The opposite effect may happen. If the participants are uninterested
What are Demand Characteristics?
This is when participants form an idea about the purpose of the study. If they guess what result is expected, they may try to please the researcher and the results will be inaccurate (or the opposite may happen)
What is Social Desirability bias?
People usually try to show themselves in the best light. So they may not be truthful but give answers that are more socially acceptable . This would make the results less accurate.
Researchers can affect the outcomes of observations in undesirable ways. Name two effects that show this.
Researcher (or Experimenter) bias - researchers expectations can influence how they design the study and how they behave towards the participants and thus produce demand characteristics. Their expectations may also influence how they take measurements and analyse data.
Interviewer Effects - interviewers expectations may lead to only asking questions that they re interested in. Or ask leading questions. Or focus on aspects that fit in with their expectations. Their behaviour and appearance can also influence participants.
What are the 4 main ethical issues to consider when doing research?
Deception
Informed consent
Psychological harm
Confidentiality
What are the BPS guidelines on Deception?
Deception should be avoided wherever possible and only be used where it is scientifically justified - when the study would be meaningless otherwise.
Deception shouldn’t be used if its likely that the participant will be unhappy when they discover the study’s true nature
If a researcher does use deception, what should they do immediately afterwards?
They should reveal the true nature of the study to the participant and give them the right to withdraw the data you collected on them during the study
What are the BPS guidelines on Informed Consent?
Participants should be given all the information they need to decide whether to participate in the research and shouldn’t be coerced or pressured
Some people may not be able to give real informed consent - for example children. In this case informed consent should be obtained from parents or guardians.
What are the BPS guidelines on Psychological Harm?
Researchers have the responsibility to protect the participants from physical and psychological harm during the study. Any risk of harm should be no greater than what the participant may experience in their normal life.
How do researchers keep confidentiality?
- Sensitive information, results or behaviour revealed through research will not be shared with others
- Information revealed through the research will not be shared unless the participant gives consent
- No information should be identifiable to an individual
- Participants should remain anonymous
What are the arguments for and against animal research?
Support - valuable information gained and some research designs couldn’t be used on humans (like Harlow’s baby monkeys)
Against - some argue that is ethically wrong
Do BPS ethical guidelines solve all problems?
Nope - psychologists who break the guidelines can’t be banned from practice - just from the BPS
How do you assess psychological harm? How do you assess whether deception was justified? How do you assess whether the end results justified the means? Difficult issues, all.
What are the problems with data analysis from quantitive research?
Adequate data sampling
Representative population
Good Operationalisation
Researcher bias
What are the advantages and disadvantages of quantifying qualitative data?
Can be limited use so need to translate it into quantitive data
The advantages of quantifying data
Easier to see patterns
Easier to summarise
Can do statistical analysis
The disadvantages of quantifying data are
Care is needed to avoid bias
Qualitative data is more detailed and this may be lost
What are the 3 measures of central tendency?
The Mean
The Median
The Mode
What are the advantages and disadvantages of using the Mean average?
Advantages
It uses all the scores in a data set
It is used to do further data analysis - standard deviation
Disadvantages
It can be distorted by outlier values (extreme high or low values)
It can give an unrealistic value as a result ( e.g. 2.4 children in a typical family)
What are the advantages and disadvantages of using the Median average?
Advantages
It’s quick and easy to calculate
It’s. not affected by extreme high or low values
Disadvantages
Not all the values are used to work it out
It has little further use in data analysis
What are the advantages and disadvantages of using the Mode?
Advantages
It shows the most common or important score
It’s always a realistic figure - not 2.4!
Disadvantages
Not very useful if there are several modal values or if the modal value is only slightly more common
It has little further use in data analysis
What do Measures of Dispersion tell you?
Measures of Dispersion tell you how spread out the data is
What are the Measures of Dispersion?
Range. Particularly Interquartile range
Standard deviation
How do you work out Mean?
Add up all the values and divide by the number of values
How do you work out the Median?
Put all the values in order and find the middle value
How do you work out the Mode?
The mode is the most often occurring value
How do you work out the Range?
Highest value - lowest value
What is the advantage and disadvantage of using the Range?
Advantage
It’s quick and easy
Disadvantage
It can be misleading if there are very high or low values
How do you work out the Interquartile Range (IQR)?
IQR = Q3 - Q1
What does Standard Deviation measure?
Standard Deviation measures how much values deviate from the Mean
What are the advantages and disadvantages of using Standard Deviation?
Advantages
All scores in the set are taken into account, so it’s more accurate than the range.
It can be used in further analysis
Disadvantages
It not quick or easy to calculate
What does Correlation measure?
Correlation measures how closely two variables are related
What is the correlation coefficient?
A number between -1 and +1
If the coefficient is close to -1 or +1 then it is strongly related a smaller number means a weak relationship
What does a positive correlation coefficient mean?
That the variables rise and fall together
What does a negative correlation coefficient mean?
That one variable rises as the other falls
What does a correlation coefficient of zero indicate?
That there is no correlation
What are the advantages of correlational research?
- Because correlational research doesn’t involve controlling any variables, you can do it when (for practical or ethical reasons) you couldn’t do a controlled experiment - like correlation between smoking and cancer
- It can give ideas for future research
- Can be used to test reliability and validity
What are the limitations of correlational research?
- It can’t establish cause and effect relationships - only a controlled experiment can do this
- Care must be taken when interpreting correlation coefficients - high coefficients could be down to chance. To decide whether a coefficient is significant, you have to use a proper significance test.
What is Spearman’s Rho?
Spearman’s Rho is a Correlation Coefficient
Name some ways to present quantitive data
Tables
Bar charts (for non-continuous data)
Histograms (for continuous data)
Frequency Polygons (showing more than one set of data)