Research methods Flashcards
What is a laboratory experimental method?
And evaluate
An experiment where the environment is set up and highly controlled. Must be a random assignment of participants to conditions. IV and DV is manipulated
High in internal validity
Low in ecological validity
What is a field experimental method?
And evaluate
An experiment where the IV and DV are manipulated however it still differs to a lab due to the setting being a real life setting (often are unaware in experiment)
High in ecological validity
Ethical issues (need informed consent and right to withdraw), extraneous variables
What is a quasi experimental method?
And evaluate
An experiment where IV is naturally occurring however there is no random allocation of condition to participants. So participants are allocated to something like age.
Allows sensitive/ unethical events to be researched, comparisons between types of people
Lacks internal validity, confounding variables
What is a natural experimental method?
And evaluate
An experiment where something has naturally occurred and a researcher decides to investigate it. IV is naturally occurring.
Allows sensitive/ unethical events to be researched, high ecological validity
No control over extraneous variables
What is an independent groups design?
Evaluate
Participants are split into groups where each do only one condition of the IV. Used in quasi and natural
Less likely to guess the purpose, no order effects
Need a large sample, can’t control for participant variables Controlled via random allocation
What is a repeated measures design?
Evaluate
Each participant does both the conditions
Don’t need a larger sample, no participant variables
Effected by order effects (counterbalancing), demand characteristics
What is a matched pairs design?
Evaluate
Participants do one condition yet they’re separated into groups based on similarities so they act as a control for each other
Less participant variables, no order effects
No two people are the exact same, lots of time
What are the measures of central tendency?
(aim to show central position of a data set)
mean (most sensitive, distorted by extreme scores)
median (unaffected by extreme scores, can be misleading not all scores taken into account)
mode (unaffected by extreme scores, an unrepresentative measure doesn’t tell us anything about other scores)
What are the measures of dispersion?
(aim to show the spread of scores in a data set)
Range (easy to calculate, can be distorted by extreme scores)
Standard deviation- how close the scores are to the mean (takes into account all scores, may hide some extreme values)
What is ecological validity?
How well does the setting represent real life
What is temporal validity?
How well do the findings represent todays view
What is population validity?
How well do the findings represent the population as a whole
What is mundane realism?
How well does the task represent something you’d do in every day life
What is internal validity?
Is the experiment well controlled and well designed ?
What are demand characteristics?
When a participant guesses the purpose of a study and as a result acts unnaturally.
What is an extraneous variable?
Any variable which is not the independent variable which has effected the results.
Name the six BPS ethics guidelines
Informed consent Lack of deception Confidentiality A right to withdraw Protection from physical and psychological harm Privacy
What are the types of consent?
Presumptive- asking a similar group of people
Prior general consent- the group agree to a range of studies including the one they want to take part in
What are the moral principles?
Respect
Competence
Responsibility
Integrity
What is random sampling?
Every member of the population has a chance of being selected.
Best chance of being unbiased and representative
But is time consuming and not everyone would be willing
What is opportunity sampling?
Sample of people who are available at the time of asking.
Quick and convenient
But may be unrepresentative
What is volunteer sampling?
Participants join a study in response to posters or adverts etc
Ethical and not much attrition
Could be unrepresentative due to only willing people
What is systematic sampling?
Participants are selected at fixed intervals from a list
Not biased
Time consuming
What is a stratified sample?
The population is classified into categories and a specific amount from each is selected
Most representative of population
Very time consuming
What is snowball sampling?
By using word of mouth
Sampling is easier for sensitive subjects
Prone to bias
What do correlations do?
Aims to discover a relationship between two co variables
What is a negative and a positive correlation?
Positive- as one increases the other one does also
Negative- as one increases the other decreases
What do the following mean about a correlation…
-1, 0, +1
-1 = a perfect negative correlation 0 = no correlation at all \+1 = a perfect positive correlation
What are the ways of assessing validity?
Face validity (look at the research tool and check the questions are focused and measuring what they're meant to) Concurrent validity (already established questionnaire compare results)
How is validity improved?
Use of a control group
Doing a single blind
Doing a double blind
Use a lie scale
What are the ways of assessing reliability?
Test retest
Inter observer reliability
Inter rater reliability
How is reliability improved?
In an interview a structure can be used
In an experiment the IV and DV can be operationalised
In an observation the behaviour categories can be operationalised
What are all the research methods?
Meta analysis Case study Experiment Self report Correlation Observation
When would you use a bar chart?
Non continuous data, observation
When would you use a histogram?
Continuous data
When would you use a scattergram?
Correlations, to show a relationship between variables
Which hypothesis do we accept when the data is significant?
The alternative hypothesis, the non directional or directional
Which hypothesis do we accept when the data is not significant?
The null hypothesis
What is a type 1 and type 2 error?
Type 1 = claims there is significance but there is not
Type 2 = claims there is no significance when there is
What are the three R’s when working with animals?
Reduction
Replacement
Refinement
What are the features of science?
THE PROF
Theory construction Hypothesis testing Empirical evidence Paradigm shifts Replicability Objectivity Falsifiability
What is meant by empirical methods?
Information is gained by experiments, observations, etc. We only know things from direct testing
What is meant by objectivity?
Scientist need to work free of bias from both themselves and researchers. Conditions need to be carefully controlled
What is meant by replicability?
Valid research need to show similar results when replicated.
What is meant by a paradigm shift?
A paradigm is common shared assumptions and methods within scientific discipline. A shift in one occurs when research or other factors cause a change in those common assumptions.
What is meant by falsifiability?
A theory which is not able to be proved as wrong is not scientifically correct
What is meant by theory construction?
A theory is facts compiled into a series of principles which are then used to predict and understand future research.
What is meant by hypothesis testing?
This refers to how theories are modified by testing what is expected
What is the difference between inductive and deductive?
Inductive- a hypothesis is made from observation, the hypothesis is tested and a theory is made from conclusions of testing
Deductive- a theory is made from observations, from this theory a hypothesis is made and tested. The theory is either correct of not correct
How would a content analysis be conducted?
A type of media would be obtained eg. a recorded interview
Coding units would be decided before (top down) or after (bottom up) watching
The media is watched several times
The coding units are marked each time they occur
To assess reliability, inter rater reliability and spearman’s rho
What is an overt and covert observation?
Overt = open, the person is aware they're being observed (ethical but demand characteristics ) Covert = closed, the person is unaware they're being watched (no demand characteristics but unethical)
What is a controlled and naturalistic observation?
Controlled = the psychologist observes behaviour in response to a set up situation (lacks ecological validity but same experience each time) Naturalistic = the psychologist observes behaviour in the real world (not ethical but no demand characteristics)
What is a participant and non participant observation?
Participant = observes behaviour from within a group (can see another viewpoint but often dangerous)
Non participant = psychologist observes from a distance (safer but not a better understanding)
What are the two types of observation sampling?
Event sampling = counting the number of times a certain behaviour occurs
Time sampling = recording behaviours at certain intervals
What is quantitive data and qualitative?
Quantitive = data which is numbers (easy to produce large amounts, little information) Qualitative = data which is words (detailed data, no generalisations can be made)
What is and independent and dependent variable?
Independent = a variable which is manipulated and change by the experimenter Dependent = a variable which is being measured by the experimenter
What is a case study?
In depth investigations of a single person or group of people allowing information to be gained on a specific topic
What is a closed question?
Questions only allowing answers which go into a specific category. Producing nominal data and some cases ordinal data.
Good because the questions are standardised however they lack detail
What is an open question?
Questions allowing people to express their answer in more detail, in own words.
Good because rich qualitative data is produced however it’s time consuming.
Evaluate self report techniques
Participants can be asked specific things to aim to get a better answer and situations can be hypothetical
This is only useful when the participant is willing to disclose and they may answer to please the experimenter
Name some self report techniques
Questionnaires
Interviews
Diary entries
What is an structured interview?
A structured interview is a quantitative research method where the interviewer a set of prepared closed-ended questions
Easy to replicate, quick to conduct
But not flexible and answers lack detail
What is a structured interview?
Unstructured interviews do not use any set questions, instead, the interviewer asks open-ended questions
More flexible, produce qualitative data
But are tie consuming, expensive
What are the types of questions in a questionnaire?
Closed- standardized, data easily turned quantitive and they lack detail
Open- rich qualitative data and can be time consuming and tricky to analyse
What are the factors making an experiment well designed?
Aims Length Pilot Question order Terminology Presentation
What is a pilot study?
Initial run through of an investigation which ensures all procedures and measures are correct.
Describe the peer review process
The research is sent to individuals who are experts on the topic they evaluate methods and designs used, the validity and the structure and language. They decide if the research is accepted or rejected.
What is counterbalancing and when is it used?
Counterbalancing is used in a repeated measures design. This is to correct any potential order effects that may have occurred, like the practise effect or boredom or fatigue.
What is random allocation and when would you use it?
Random allocation is used to spread participant variables across all conditions of an experiment. This is used in an independent groups design.
What are standardised procedures and when are they used?
Instructions with a clear aim to ensure all participants are tested in the exact same way. These are used to prevent situational variables from influencing the dependant variable.
What is a double blind study and when is it used?
A double blind study is when a study has an investigator who doesn’t know the purpose and also the participants don’t know the purpose. This is used to prevent investigator effects.
What is nominal data?
Data which is under categories
What is ordinal data?
Data with no specific measurement
What is interval/ ratio data?
Data with specific units or measure
What does a negatively skewed graph look like?
Graph curves to the left
What does a positively skewed graph look like?
Graph curves to the right
