Methods P2 Flashcards
Independent variable
What is manipulated by the researcher or changes natrually to test its effect on the dependant variable
Dependant variable
What is measured by the researcher and influenced by the independent variable
Extraneous variable
Everything that could effect the DV that’s not the IV
- they do vary sustematically with the IV
Confounding Variable
Any variable, other than the IV, that may have effected the DV, so we cannot be sure of the true source of changes to the DV. Confounding variables vary systematically with the IV
3 examples of extraneous variable
• participant variable
• situation variable
• investigator effects
When do you use non-directional hypothesis?
When there is no previous research
Aim
A general statement of what the researcher intends to investigate
To investigate weather…
Hypothesis
A clear, precise, testable statement that states the relationship between the variables to be tested. Stated at the beginning of the study
Direct hypothesis
States the direction of the differences or the relationship between tested groups
Non-direct hypothesis
Does not state the direction of the relationship
Operationalisation
Clearly defining variables in terms of how they can be measured
Confounding variable
A type of EV accept it varies systematically with the IV. Therefore one doesn’t know if change in DV is IV of confounding variable
Demand characteristics
Any cue from the researcher or the research situation that may be interpreted by the participant as revelling the purpose of the investigation. This may lead to a participant changing their behaviour within the research situation
Investigator effect
Any effect from the investigator’s behaviour on the DV, could be anything from dessign of study, selection and interation with particpants
Randomisation
Chance methods to to control effect of bias
Standardisation
Using same procedures for all participants
Different types of validity
External validity
Internal validity
Ecological
Temporal
Population
Different types of observations
• naturalist
• controlled
• covert
• overt
• participant
• non participant
BPS code of ethics
• respect
• competence
• responsibility
• integrity
Experimental group desing (3 + 1)
- independent group design
- repeated measure
- matched pair desing, random allocation
- counter blaencing
Independent groups
Different groups of participants in different conditions
Repeated measure
The same group of participants go through all different conditions
Matched pairs
Screen participants in order to match them in pairs with similar attributes that are being tested/ have and effect on the DV, split the pair to experience the two different conditions
Counter balancing
splitting the group in half so one group experince the condition s in one order and the other half in the other order
- attemt to control order effects
BPS code of ethics
A quasi-legal document instructing psychologists in the UK about acceptable behaviour in experiments based around respect competence responsibility and integrity
5 different types of sampling
• random sample
• systematic sample
• stratified sample
• opportunity sample
• Volunteer sample
Random sampling
All members of target population have equal chance of being selected
Systematic sampling
Every nth member of target population is selected
Stratified sampling
Composition of the sample reflects the proportion of people in certain subgroups
Opportunity sampling
Takes anyone willing and available
Volunteer sampling
Participants select themselves i.e in a response to an add
Different types of experiments (4)
• lab experiment (in the lab)
• field experiment (in natural setting control over IV)
• natural experiment (IV would have happens without researcher)
• quasi-experiment (variables exist i.e they are old)
Laboratory experiment
- takes place in an controlled environment
- the researcher manipulates the IV and records the effects on the DV
- maintaining strict control of extraneous variables
Field experiment
In the natrual environment of the participant like the street
Natural experiment
When the IV would have happened with or without the researcher
Quasi experiment
Where the variables just exist already like age
Correlation
An association between two co variables
Co variables
Unlike DV or IV co variable aren’t trying to show a cause and effect relationship
Curvilinear relationship
Meta analysis
The process of combining findings from a number of studies in a particular topic to produce an overall statistical conclusion
Ethical evaluation
Deception
Consent
Protection from harm
Difference between correlation and experiment
Correlation is not causal relationship ?
Naturalistic / controlled observation
Naturalistic is in the natural environment of participant whist controlled is in artificial setting
Covert / overt observation
Covert is observing participants without consent in a public setting whilst overt observation is when a participant has given informed consent
Participant / non-participant observations
Participant is when the observer becomes a part of the group whilst non participation observation is when the researcher remains separate to the group
Random allocation
Randomly assigned rolls to avoid bias
Evaluate structured interview
Evaluate unstructured interview
Evaluate semi-structured interview
Pilot study
A small scale version of an investigation that takes place beforehand to check that the procedures, materials and instructions work. It also allows the researchers to make any necessary changes
Single blind and double blind procedures
Single blind - participants don’t know the aim or which condition of the experiment their in
Double blind - both the researcher and participant don’t know
Measure of central tendency
Mean
Median
Mode
How do you find the mean, median, mode?
Measure of dispersion
Range
Standard deviation
What is standard deviation
Scattergram
Represents the strength and relationship between co variables in a correlation analysis
Bar chart
Frequency represented by height of bars
Histogram
Where the area of the bars represent the frequency, X axis starts at zero and is continuous
Effect size
Overall statistical measure of relationship across variables in a meta-analysis
Meta analysis
A number of research studies that have investigated the same area of review
Quantitive and Qualitative data
Quantitive - numbers
Qualitative - expressed in words
Sampling methods in observational design
event sampling - counting the number of times a particular event happened
time sampling - recording behaviour in a pre established time frame, what is the subject doing at every 30 seconds
What do behavioural catorgires mean?
Nominal data
Data represented by categories
Ordinal data
Data ordered in some way, for example rate out of 10, this lacks precision and is subjective
Interval data
Based on numerical scales that include units of equal precisely defined data
Interval data
Based on numerical scales that include units of equal precisely defined data
3 pieces of information you need for a sign test
- Need to be looking for a difference not an association
- Need to use a reappeared measure design
- Need to be using nominal data
When is the results of a sign test significant
The calculated value need to be the same or bigger than the critical value
3 things you need to find the critical value
- Significance level (usually 0.05 unless human cost or a one time thing)
- No. of participants
- Wether hypothesis is one tailed (directional) or two tailed (non - directional)
what is the point of a statistical test?
to find out is a difference or association is significant
what three pieces of information do you need to determine which statistical test to do?
- is it a differnce or correlation
- if it is a differnce what experimental design is used
- the level of mesurment (nominal, ordinal or interval)
mnemonic used to remember all the statisitcal tests
carrots should come
mashed with swede
under roast potatoes
carrots should come mashed with swede under roast potato
chi squared, sign test, chi squared
man whitney, wilcoxon, spearman’s rho
unrelated t test, related t test, pearson’s r
the speficic passge of writing you should use when declareing which statisitcal test to use
ugh
one tailed and two tailed hypotheses
One tailed, directional
Two tailed, non directional
Correlation coefficient
Number between -1 and +1 that indicates the strength and direction of correlation
Case studies
• investigation into a single group or individual
• use a combination of data collection methods
Internal validity
Did the researcher measure what they intended to measure, was the change in the DV due to the IV and not another factor such as demand characterises
External validity
Ecological validity (other settings)
Population validity etc
Face validity
Does it appear to measure what it intends to measure?
Concurrent validity
The extent to which a psychological measure relates to an existing similar measure.
The threshold correlation between two sets of scores of the same variable to be considered valid
+ .80
Validity
The extent the experiment measures what it intends to mesure
Way of assessing validity
Concurrent and face validity are measures of validity
Improving validity
• control groups
• standardisation
• single blind and double blind procedures
• lie scale in questionairs
• covert observation
• qulaitavite lees open to interpretation
Test retest
To test for reliability by assessing the same person on the the same test on two separate occasions
Inter observer reliability
The extent of the agreement between two or more observers, this is measured using correlation. No of agreements / no of observation > +.80 then the data has high inter observer reliability
How to do a sign test
How to do Spearmen’s rho
- Rank in table lowest to highest
- Calculate the difference and then square in each pair
- Put into rho equation
- Compare with critical value
Content analysis
Observational research where pp are studied indirectly via communications they have produced
Coding in content analysis
The initial stage catorgrising large amounts of data into meaningful units to produce quantative data
Thematic analysis
Any idea that keeps cropping up
Steps of content analysis
What is the null hypothesis
Opposite to the alternative hypothesis that stars there will be no difference between the different conditions
What you need to know to use the table of Cristal values
• is it one tailed or two tailed
• number of participants in study
• level of significance (usually 0.05)
Type 1 error
Incorrectly accepting the alternative hypothesis and rejecting the bill hypothesis
Type 2 error
Incorrectly rejecting the alternative hypothesis and accepting the null hypothesis
Sections of a scientific report
- abstract
- introduction
- method
- result
- discussion
- referencing
How to do wilcoxen test
Paradigm
Paradigm shift
Falsfiflablity
evaluation of qualative data
evaluation of quantitive data
