Research Methods: Key Terms Flashcards
Case Studies
an in depth investigation often of a single individual, event or small group
Interval Data
numerical data that is ordered and in objective units
Ordinal Data
numerical data that is ordered but subjective, difference between items need not be the same
Nominal Data
data separated into categories
Standard Deviation
measure of spread of data around mean, calculates average distance from mean of all scores
Correlation Coefficients
a numerical representation of a correlation, range from -1 (strong negative correlation) to +1 (strong positive correlation)
Covariables
the variables that are measured in a correlation
Correlation
the measurement of a relationship between 2 or more variables shows a relationship not the causation
Interviewer Bias
where expectations or opinions of interviewer interferes with judgement of interviewee
Self Report Techniques
research methods in which ps give info about themselves (without researcher interference)
Social Desirability Bias
participants answer qs in a manner that will be viewed favourably by others
Structured Interview
made up of pre determined set of qs asked in fixed order, standardised
Investigator Effects
when the investigator directly or indirectly has an effect on a ps performance
Demand Characteristics
when a subject picks up cues during an experiment therefore possibly affecting + altering the results
Mundane Realism
whether or not task ps are asked to do represent something they would experience and do in real life
Internal Validity
concerns things that go on whithin a study that may affect accuracy
Ecological Validity
concerns whether results can be generalised to other settings
Temporal Validity
concerns whether results can be generalised to other times
Population Validity
concerns whether results can be generalised to other people
External Validity
concerns whether results can be generalised to different situations
Quasi Experiment
IV is naturally occuring + is impossible to manipulate, cant be randomly assigned to experimental + control groups
Natural Experiment
conducted when not possible, ethically or practically, to manipulate IV. DV can be tested in a lab
Field Experiment
controlled experiment conducted outside a lab, IV is manipulated but difficult to control extraneous + confounding variables, ps unaware, cause + effect inferred
Lab Experiment
an experiment carried out in a controlled setting, ps are aware, variables carefully controlled, artificial materials, cause + effect are established
Volunteer Sampling
advertise in newspaper/notice board or internet, variety of ps, volunteer bias: have more time, are helpful, need money
Systematic Sampling
using predetermined system to select participants, same number has to be applied consistently unbiased, only truly unbiased if use random method to come up with number
Stratified Sampling
subgroups are identified within a population and participants are obtained from each strata in proportion to their occurence in the population, selection using random, representative, time consuming
Opportunity Sampling
select people who are most easily available, easy, biased
Random Sampling
random technique, every member of target population has equal chance, may take time
Confounding Variable
variables that arent IV but varies systematically with IV, can stop from establishing cause + effect
Extraneous Variable
variables that change other than IV and are quite difficult to control, they make it difficult to detect a significant effect
Pilot Study
small scale practice run of study to identify problems in design, method or analysis
Standardisation
everything said must be scripted in advance to ensure fairness across conditions
Matched Pairs Design
two equal groups used, one for each condition with the ps matched based on key variables
Repeated Measures Design
only one group of ps is used for both conditions of the experiment
Independent Groups Design
a separate group of ps for each condition of the experiment
Experimental Design
how the ps are picked to stop other things interfering with results
Null Hypothesis
states that there will be no difference between the 2 conditions in the experiment
Non Directional Hypothesis
a hypothesis that states there would be a difference between the 2 conditions but wouldnt specify the direction it would go
Directional Hypothesis
a hypothesis that indicates which direction the results will go
Operationalism
strictly defining the variables into measurable factors
Dependent Variable (DV)
what you measure in an experiment
Independent Variable (IV)
what you change in an experiment
Aim
starts with ‘to investigate/to establish’
Cost Benefit Analysis
a balance between the best interests of the ps and the value of the research
Controlled Observation
behaviour is observed under conditions where certain variables have been organised by the researcher
Naturalistic Observation
it takes place in an everyday setting with no interference
Overt Observation
the ps are aware they are being studied
Covert Observation
it is kept secret from the ps that they are being studied
Participant Observation
the observations are being made by someone who is also participating in the activity being observed
Non-Participant Observation
the observer is separate from the people being observed
Reporting Psychological Investigations: Title
- concise and informative
- contains variables and researcher names
Reporting Psychological Investigations: Abstract
- summary (150-200 words) of the entire report
- first thing in the report
- some detail of the aims, procedure, findings and conslusion
Reporting Psychological Investigations: Introduction
- review of background studies in the same area
- includes aims and hypothesis
- short essay
Reporting Psychological Investigations: Method
- mention experimental design, population, sample, sampling method, equipment, procedure, ethics
Reporting Psychological Investigations: Results
- contains descriptive and inferential statistics
Reporting Psychological Investigations: Discussion
- summary of the results and small explanation
- relationship of results to background research
- evaluation of the investigation (methodological critique)
- suggestions for future research
- implications for psychological theory and real life
Reporting Psychological Investigations: References
- list of all the sources and where to find original articles
Reporting Psychological Investigations: Appendices
- material that would overload the report eg raw data or research tool
Descriptive Statistics
provide overview of trends in the form of a graph, table or chart
Inferential Statistics
show if the results obtained are significant eg sign test
Meta-Analysis
a researcher looks at the findings from a number of different studies and produces a statistic to represent the overall effect
Content Analysis
a kind of observational study in which behaviour is observed indirectly in written or verbal material such as interviews, conversations, books, diaries or TV programmes
Thematic Analysis
identifying emerging facts that we identify during ad after analysing the data
p Value
- the probability that the obtained results are due to chance
- anywhere between 0 and 1
- 1 means it’s very likely that they were down to chance
- 0 means that it’s unlikely
What to do with the null hypothesis when you have p
- if p is less than the significance level you reject the null hypothesis
- if p is more than the significance level you accept the null hypothesis
Type I Error
- a false positive
- falsely rejecting the null hypothesis
- the probablity of the type I error is the significance level
Type II Error
- a false negative
- falsely accepting the null hypothesis
4 Stages of Statistical Testing
1) set a significance level
2) obtain data
3) calculate p
4) reject/accept null hypothesis
The Normal Significance Level and Why
- 5%
- it balances the probability of a type I error and a type II error occurring
- 95% sure the results didn’t happen by chance
What Happens When Significance Level is Too Low
- test is too stringent
- too difficult to acheive significance
- overlooks true effects because they didn’t achieve significance
What Happens When Significance Level is Too High
- test is too lenient
- too easy to achieve significance
- detects true differences where in actual fact the results are due to chance
Face Validity
- an intuitive assessment of whether the test actually measures what it sets out to measure
- the degree to which an experiment assesses the aims stated at the beginning
Concurrent Validity
- whether or not a test correlates with a previously validated test
- the extent of which 2 experiments agree with eachother
- must have a correlation of above +0.8 to have high concurrent validity
Lie Scale
- asks the same question in different ways at different times in the questionnaire to see whether a participants answers are consistent
- if they are consistent then it’s valid
Interpretive Validity
- ask the participants if they agree with the interpretation of the outcome/results made by the researcher
- if they agree then it’s valid
Purpose of Inferential Statistics
allow us to make inference/decision about whether to accept or reject the null hypothesis
Objectivity
not affected by expectations or bias
Empirical Method
method of investigation that relies on direct observation or testing
Replicability
the repeatability of a procedure and/or outcome of a study
Falsifiability
the possibility that a statement or hypothesis can be proven wrong
Theory Construction
a collection of general principles that explain observations and facts
Hypothesis Testing
the process of testing the validity of a theory by testing the predictions that it makes
Paradigms and Paradigm Shifts
a shared set of assumptions about the subject matter of a discipline and the methods approprtiate to its study
7 Features of Science
1) objectivity
2) empirical method
3) replicability
4) falsifiability
5) theory construction
6) hypothesis testing
7) paradigms and paradigm shifts
