PAPER 2 - Research Methods Flashcards
What is an INDEPEDENT VARIABLE ?
variable that is MANIPULATED by the researcher
What is a DEPENDENT VARIABLE ?
the variable that is MEASURED
E.G: reaction time in a driving stimulator
What are EXTRANEOUS VARIABLES ?
a variable the might affect your DV - they are identified before the study
Give an example of an extraneous variable
room temp, time of day, task given
What are CONFOUNDING VARIABLES ?
type of extraneous variable that isn’t controlled and affect the DV
change systematically with IV.
What is the OPERATIONALISATION of variables ?
how the variables are made measurable - drawing out the most important elements of the variables
Give an example of the operationalisation of variables
intelligence is a very broad term - to make it measurable we could use a specific intelligence test that measures certain elements of personality
What are DEMAND CHARACTERISTICS ?
when the pp works out how the researcher wants them to behave - can be conscious or unconscious - reduces internal validity
Name 2 ways of reducing demand characteristics
- use different pp in each condition (INDEPENDENT GROUPS)
- use SINGLE BLIND TECHNIQUE where the pp does not know which condition of the experiment they are assigned to
What are INVESTIGATOR EFFECTS ?
when the researcher influences the results
What are EXPECTATION EFFECTS ?
when a researcher is DEEPLY COMMITTED to achieving a particular outcome
Name one method of reducing investigator effects
DOUBLE BLING TECHNIQUE - where neither the pp or research know the aim and/or condition of the study
What are the 3 experimental methods ?
lab / field / natural
What are the 3 key features of a lab method ?
- direct manipulation of IV
- control
- randomisation
What are the STRENGTHS of a lab method ?
EASY REPLICATED - high control - able to check reliability
INTERNAL VALIDITY - easier to control confounding variables - able to establish cause and effect
What are the LIMITATIONS of a lab method ?
ECOLOGICAL VALIDITY - high levels of control - generalisability
DEMAND CHARACTERISTICS - pp know they are being researched - reduces internal validity
What are the key features of a field method ?
- natural environment
- still an IV and DV
- attempt to control extraneous variables
- cause and effect can be established because pp is unaware they are observed
What are the STRENGTHS of a field method ?
ECOLOGICAL VALIDITY - behaviour is representative of everyday life - generalisable
DEMAND CHARACTERISTICS - do not know they are taking part - increases internal validity
What are the LIMITATIONS of a field method
TIME CONSUMING - small amount of people at certain times - reduced population validity
CONTROL - may be uncontrolled extraneous variables - reduces internal validity
What are the key features of a natural method ?
- no control over IV
- pp already assigned to a condition of the IV
What are the STRENGTHS of a natural method ?
ECOLOGICAL VALIDITY - generalise
DEMAND CHARACTERISTICS - increases the internal validity
What are the LIMITATIONS of a natural method ?
REPLICABLE - internal validity
CONTROL - not possible to predict everything - internal validity
What is a QUASI EXPERIMENT
use a pre-existing IV - one that exists e.g. character trait such as gender / age
What are the 3 types of experimental design ?
repeated measure
independent groups
matched pairs
What is a REPEATED MEASURE design ?
all the pp take part in each condition
What are the STRENGTHS of a repeated measure design ?
LESS PPS NEEDED
PPS VARIABLES AREN’T AN ISSUE
What are the LIMITATIONS of a repeated measure design ?
ORDER EFFECTS
DEMAND CHARACTERISTICS
What is an INDEPENDENT GROUPS design ?
different pps take part in each condition
What are the STRENGTHS of an independent groups design ?
ORDER EFFECTS
DEMAND CHARACTERISTICS
What are the LIMITATIONS of an independent groups design ?
PPS VARIABLES
LOST OF PPS NEEDED
What is a MATCHED PAIRS design ?
different pps take part in each condition but they are matched on characteristics e.g. gender / age
What are the STRENGTHS of a matched pairs design ?
PPS VARIABLES
DEMAND CHARACTERISTICS
What are the LIMITATIONS of a matched pairs design ?
TIME CONSUMING
LOTS OF PPS NEEDED
How could you deal with the limitations of a repeated measures design ?
COUNTERBALANCING - complete conditions in different order to BALANCE OUT any differences
How could you deal with the limitations of an independent groups design ?
RANDOMLY ALLOCATE to condition
How could you deal with the limitations of a matched pairs design ?
PILOT STUDY - consider variables that could be used to match pps
What is the FATIGUE EFFECT ?
deterioration of performance across condition as the pp becomes tired or bored
What is the PRACTICE EFFECT ?
improvement across condition through familiarity of the task or environment
What is RANDOM ALLOCATION ?
(independent groups) - control pp variables - same chance of being allocated to either condition - names in a hat
What is COUNTERBALANCING ?
(repeated measure) - complete conditions in different order - balance out differences among pps
What is RANDOMISATION ?
presenting stimuli in a random order to avoid bias - methods may include tossing a coin
What is STANDARDISATION ?
pps have EXACTLY THE SAME EXPERIENCE - methods may include written instructions
What is a NATURALISTIC OBSERVATION ?
observing naturally occurring behaviours - natural setting
What are the STRENGTHS of a naturalistic observation ?
EXTERNAL VALIDITY - ecological validity
What are the LIMITATIONS of a naturalistic observation ?
EXTRANEOUS VARIABLES
What is a CONTROLLED OBSERVATION ?
researcher attempts to control certain variables - pps know they are being observed
What are the STRENGTHS of a controlled observation ?
CONTROL - able to replicate
What are the LIMITATIONS of a controlled observation ?
EXTERNAL VALIDITY
What is an OVERT OBSERVATION ?
in the open - pps aware they are being observed, nature + purpose
What are the STRENGTHS of an overt observation ?
ETHICS - easier to replicate
What are the LIMITATION of an overt observation ?
DEMAND CHARACTERISTICS - reduces validity
What is a COVERT OBSERVATION ?
pps do not know they are being observed
What are the STRENGTHS of a covert observation ?
DEMAND CHARACTERISTICS - increases validity
What are the LIMITATIONS of a covert observation ?
ETHICS
What is a PARTICIPANT OBSERVATION ?
observer joins in
What are the STRENGTHS of a participant observation ?
UNDERSTAND BEHAVIOUR - increases validity
What are the LIMITATIONS of a participant observation ?
DEVELOP RAPPORT - reduces objectivity
What is a NON-PARTICIPANT OBSERVATION ?
researcher remains external
What are the STRENGTHS of a non-participant observation ?
OBJECTIVE - no investigator effects
What are the LIMITATIONS of a non-participant observation ?
VALUABLE DATA IS MISSED
What are BEHAVIOURAL CATEGORIES ?
how researcher operationalises - researcher knows what constitutes the behaviour being observed
Give an example of a behavioural category
if you were observing AGGRESSIVE behaviour, the researcher would need to know what constitutes as aggressive such as kicking, punching
What are the two SAMPLING PROCEDURES ?
event sampling
time sampling
What is EVENT SAMPLING ?
watching the event and recording every time a specific behaviour occurs
What is TIME SAMPLING ?
recording behaviour at set time intervals
What are the two self-report techniques ?
questionnaire
interview
What is a SELF-REPORT technique ?
allows pps to PROVIDE INFORMATION about specific thing relating to themselves
What is a QUESTIONNAIRE ?
printed series of questions used to gather opinions around a certain topic
Name methods that can be used to distribute questionnaires
post / telephone / internet
Is it important to get a large representative sample of questionnaires ? Why ?
yes - to make a generalisation
What is a CLOSED QUESTION ?
produce QUANTITATIVE data - ticking boxes / circling answers
What is an OPEN QUESTION ?
produces QUALITATIVE data - difficult to analyse - not restricted in available answers
What is a LEADING QUESTION ?
should be avoided - choice of wording suggests a certain answer should be given
What are the STRENGTHS of a questionnaire ?
SIMPLICITY - minimum of training - easy to analyse - researcher doesn’t need to be present
INTERPERSONAL VARIABLES - more likely to answer honestly - internal validity
What are the LIMITATIONS of a questionnaire ?
WORDING OF QUESTIONS - pps may interpret question wrong - leading questions may influence response - internal validity
RESPONSE RATE - 30% or less - appeal to those who like the research topic - population validity
What are STRUCTURED INTERVIEWS ?
- same questions
- same order
- useful of teams of interviewers
- comparisons can be made
What are SEMI-STRUCTURED INTERVIEWS ?
- flexible in order
- flow more naturally
- personal response
- follow-up questions
What are UNSTRUCTURED INTERVIEWS ?
- informal
- conversation about the topic
- interviewee largely in control
What are the STRENGTHS of interviews ?
ACCURATE DATA - rapport - more honest answers - ask for clarification - training reduces leading questions - internal validity
DETAILED DATA - open ended questions - clarify meaning - develop understanding
What are the LIMITATIONS of the interviews ?
HARD TO ANALYSE - qual. data - subjective - different researchers interpret data differently - reduces validity
INTERPERSONAL VARIABLES - reducing internal validity
What is a CORRELATION ?
relationship between two things
What are the correlation coefficient numbers ?
+1 perfect positive correlation
-1 perfect negative correlation
What is a POSITIVE CORRELATION ?
- as one variable increases so does the other
What is a weak positive correlation ?
r = +0.2
What is a strong positive correlation ?
r = +0.9
What is a NEGATIVE CORRELATION ?
- as one variable increases the other decreases
What is a weak negative correlation ?
r = -0.2
What is a strong negative correlation ?
r = -0.9
What is ZERO CORRELATION ?
no relationship between two variables
What figure would show zero correlation ?
r = -0.02 r = +0.06
What is the FIRST difference between an experiment and a correlation ?
E = manipulation of IV C = no deliberate change to any variable
What is the SECOND difference between an experiment and a correlation ?
E = DV is measured C = impact on one variable is not measured - comparison is made between 2
What is the THIRD difference between an experiment and a correlation ?
E = establish cause and effect C = cannot establish cause and effect
What are the STRENGTHS of correlations ?
DIRECTION - precise quantitative measure of the strength of the relationship - researcher knows whether to carry out experiment
STATISTICALLY ANALYSE - not all situations can be ethically experimented e.g. aggression
What are the LIMITATIONS of correlations ?
CAUSE AND EFFECT - may be third variable e.g. divorce is the reason a child is both aggressive and in day car - need to be careful when drawing conclusions
NONLINEAR RELATIONSHIPS - correlation might initially be positive but then become negative, the two relationships cancel each other out
What is CONTENT ANALYSIS ?
changing qualitative data into quantitative data using a coding unit (s)
What is a CODING UNIT ?
specific behaviours / words / phrases that you are going to count in the material
What is a TOP DOWN coding unit ?
decided BEFORE content analysis using pre-existing coding units
What is a BOTTOM UP coding unit ?
decided AFTER you have viewed the material
What are the STRENGTHS of content analysis ?
LARGE AMOUNTS OF QUALITATIVE DATA QUICKLY - more representative results
EASY TO TEST FOR RELIABILITY - inter-rater tests - more objective
What are the LIMITATIONS of content analysis ?
BIAS - researcher chooses coding units - reduces internal validity
MAY NOT BE ACCURATELY ANALYSED - behaviour that does not fit coding unit will be ignored - reduces internal validity
What is a CASE STUDY ?
in-depth study over time of a ‘case’ which is usually an individual or small group
What is an example of a case study ?
CLIVE WEARING
- contracted Herpes Simplex virus
- attacked the brain
- unable to process new memories
- unable to control emotions
What are the STRENGTHS of case studies ?
RICH DATA - provide valuable new data
ETHICS - privation
What are the LIMITATIONS of case studies ?
RELIABILITY - unlikely to be replicated - relies on pps memory being accurate
SUBJECTIVE - relationships may be established - biased - not be a true reflection
What is a PILOT STUDY ?
small scale study - make sure materials / procedures are suitable
What issues need to be identified in a pilot study of an experiment ?
- materials
- timings
- instructions
What issues need to be identified in a pilot study of an observation ?
- behavioural checklist
- covert position
- cctv
- inter-observer ratings
What issues need to be identified in a pilot study of an interview ?
- questions
- recording method
What are the benefits of carrying out pilot studies ?
- makes sure observations are reliable
- saves time and money
What is a TARGET POPULATION ?
- group of people who share a set of characteristics
- large
What is a SAMPLE ?
- subset of the target population
What is the aim of finding a sample ?
- representative of target population
- larger the sample the more accurate (not always)
What is an OPPORTUNITY SAMPLE ?
selecting those who are readily available
OPPORTUNITY SAMPLE : EXAMPLE
going to the common room to look for students
OPPORTUNITY SAMPLE : BIAS
researcher bias - researcher chooses sample
OPPORTUNITY SAMPLE : GENERALISABILITY
unlikely - everyone who doesn’t have science will be in the common room
What is a RANDOM SAMPLE ?
all the members of the target population have an equal chance of being selected
RANDOM SAMPLE : EXAMPLE
pulling names from a hat
RANDOM SAMPLE : BIAS
free from researcher bias
RANDOM SAMPLE : GENERALISABILITY
should be generaliseable but there is a chance you get a majority of males / females
What is a SYSTEMIC SAMPLE ?
putting pps in an order and selecting every nth from the list
SYSTEMIC SAMPLE : EXAMPLE
selecting every 5th name from the register
SYSTEMIC SAMPLE : BIAS
free from researcher bias
SYSTEMIC SAMPLE : GENERALISABILITY
should be representative but some small groups e.g. language students may be missed
What is a STRATIFIED SAMPLE ?
same proportion of people in the target population will be represented in the sample
STRATIFIED SAMPLE : EXAMPLE
Target population = 50% male / 50% female
Sample = 50% male / 50% female
STRATIFIED SAMPLE : BIAS
free from researcher bias
STRATIFIED SAMPLE : GENERALISABILITY
most generalisable because all groups will be represented
What is a VOLUNTEER SAMPLE ?
place an advert asking for volunteers
VOLUNTEER SAMPLE : EXAMPLE
puting an advert in a food magazine to attract chefs
VOLUNTEER SAMPLE : BIAS
volunteer bias - they are choosing to take part so it is likely they are all interested in the topic being researched
VOLUNTEER SAMPLE : GENERALISABILITY
unlikely - not be representative of all types of people
What is an AIM ?
- general statement
- what is being investigated
- developed from previous theories
Give an investigation of aim
an investigation into whether the level of processing involved affects whether the information is remembered
What is a HYPOTHESIS ?
- formulated after the aim
- prediction about the outcome
What is a NULL HYPOTHESIS ?
states there will no difference / correlation
When is an EXPERIMENTAL HYPOTHESIS used ?
when an experimental method has been carried out and there is an IV and a DV
When is a NON_EXPERIMENTAL HYPOTHESIS used ?
when there is no IV or DV e.g. correlation / observation
What is a DIRECTIONAL HYPOTHESIS ?
- predicts the direction
- more / less / greater
- lots of previous research
What is a NON-DIRECTIONAL HYPOTHESIS ?
- does not predict the direction
- little existing research
What is INFORMED CONSENT ?
- the investigator should inform all pps of the objectives
- inform pps of all aspects of the research
- safeguarding for those with special needs
What is DECEPTION ?
- unacceptable to withhold information
- intentional deception should be avoided where possible
What is PROTECTION FROM PARTICIPANTS ?
- pps must not be harmed (physiological or psychological)
What is RIGHT TO WITHDRAW ?
- pps can withdraw at any time
- results can be withdrawn
- should not be told they cannot withdraw
What is DEBRIEFING ?
- pps should leave as entered
- researcher should discuss their experience
- re-explain aims / objectives
What is CONFIDENTIALITY ?
- any information gained must remain confidential
- pps have the right to know their data will remain confidential
What should you write in a CONSENT FORM ?
- topic
- expected to do
- anything that might affect willingness to participate
- confidential
- withdraw
- place to sign
- place for questions
What should you write in a DEBRIEFING FORM ?
- thank you
- specific hypothesis
- what you expect to find
- withdraw
- how to contact you
- source of support
- questions
What are the methods of dealing with deception ?
- give full debriefing
- pps told what their results will be used for
- pps have right to withdraw data
What are the methods of dealing with informed consent ?
- pps should be given a consent letter
What is PRESUMPTIVE CONSENT ?
similar group of people are asked if the experiment is acceptable
What is PRIOR GENERAL CONSENT ?
pps give consent to a number of studies
What is RETROSPECTIVE CONSENT ?
pps are asked for consent (during debriefing) having already taken part
What are the methods of dealing with protection from harm ?
- pps should be reassured their behaviour is normal
- may require counselling
What are the methods of dealing with confidentiality ?
- personal data must be protected
- pps real names must not be used e.g. numbers
What is COST BENEFIT ANALYSIS ?
- researcher ways pros against cons
- may be looked at from pp point of view
What is the ETHICS COMMITTEE ?
- approve any study before it begins
- consider possible ethical issues and how the researcher could deal with them
What is a PEER REVIEW ?
- psychology papers under INDEPENDENT SCRUTINY
- consider validity, significance and originality
What is an OPEN peer review ?
both the researcher and the reviewer are known to each other
What is a SINGLE BLIND peer review ?
the reviewer known the researcher
What is a DOUBLE BLIND peer review ?
neither the researcher nor the reviewer knows who the other is
What is the PURPOSE of a peer review ?
HIGH QUALITY - highest quality of research
CONTRIBUTION - important contribution to the scientific field
GENUINE - refers to relevant research by other psychologists
What are the LIMITATIONS of a peer review ?
INTEGRITY - allow the reviewer to research the issue OR the work has already been plagiarised
BIAS TOWARDS POSITIVE FINDINGS - ‘file-drawer problem’
TIME CONSUMING - months to review and send back with suggestions, can take years if adjustments need to be made
What does RELIABILITY measure ?
consistency
What are 2 ways of measuring reliability ?
test - retest method / inter - observer reliability
What is the TEST-RETEST METHOD ?
- person is given same test / interview etc but a few weeks later (long enough so pp doesn’t remember answer)
- outcome should be similar every time
- correlation can be carried out between the scores
What is the INTER-OBSERVER METHOD ?
- checks to see each observer is interpreting the behavioural categories in the same way
- measure the extent to which DIFFERENT OBSERVERS get SIMILAR RESULTS for the SAME PPS
- sets of results are correlated
What is EXTERNAL RELIABILITY ?
extent to which a measure varies from one use to another
What is INTERNAL RELIABILITY ?
extent to which a measure is consistent with itself
How would you test the reliability of an EXPERIMENT ?
test-retest method
How could a researcher improve the reliability of an EXPERIMENT ?
- variables are operasionalised
- tight control of extraneous variables
- clearly described materials
- in-depth method section
- standardised instructions
How would you test the reliability of an OBSERVATION ?
inter-observer reliability
How would you improve the reliability of an OBSERVATION ?
- multiple observers (good training)
- coding system
- clear behavioural checklist
- filming behaviour
How would you assess the reliability of a SELF-REPORT ?
split-half method
How would you improve the reliability of a SELF-REPORT ?
- pilot study
- check questions are non-ambiguous
- trained interviewers
What is the SPLIT-HALF METHOD ?
measure the extent to which all part of the test contribute equally to what is being measured
What does VALIDITY measure ?
accuracy
What is INTERNAL VALIDITY ?
- what goes on inside the study
- sees if the researcher tested what they intended to test
What are the two methods of testing internal validity ?
- face validity
- concurrent validity
What is FACE VALIDITY ?
- tests internal validity
- whether at face value the study appears to measure what it should have done
What is CONCURRENT VALIDITY ?
- tests internal validity
- compares new measures against previously validated measures
What is EXTERNAL VALIDITY ?
- concerned with factors outside the study
- the extent to which the findings can be generalised
What are the two ways of testing external validity ?
ecological validity
temporal validity
What is TEMPORAL VALIDITY ?
- extent to which we can generalise to other time periods
e. g. is a 1950s study applicable to behaviour today ?
How can you improve internal validity in an EXPERIMENT ?
- operationalise variables
- control extraneous variables
- adapt IV or DV
How can you improve the external validity of an EXPERIMENT ?
- change sampling technique
- replication with different sample will improve population validity
- making IV and DV true to real life
How can you improve the internal validity of an OBSERVATION ?
- multiple observers
- well trained observers
- validated coding system
- check behavioural checklist
How can you improve the external validity of an OBSERVATION ?
- observe as many people as possible
- covert observation
- adjust behavioural checklist
How can you improve the internal validity of an QUESTIONNAIRE ?
- questions should be a valid representation of the behaviour being studied
How can you improve the external validity of a QUESTIONNAIRE ?
- accessible e.g. complete online
- updated to suit contemporary society
What are the 5 features of science ?
- objectivity
- theory construction
- hypothesis testing
- replication
- falsifiability
What is OBJECTIVITY ?
- expectation should not affect what they record
- stop any personal prejudice
- standardised instructions / operationalised variables / double blind technique
What is the EMPIRICAL METHOD ?
using research evidence to help develop and find support for theories
What is THEORY CONSTRUCTION ?
- using facts to construct theories
- collection of general principles
- must be testable and falsifiable
What is HYPOTHESIS TESTING ?
- testing to see if a researcher hypothesis is true or not
- formed from previous research
- researcher gathers evidence which helps to develop and inform the theory
What is the RESEARCH CYCLE ?
theory - devise hypothesis - test theory using appropriate method - analyse results and compare to hypothesis and theory
What is REPLICATION ?
- results could have been a fluke
- all details of a study need to be published
- compare results
What is FALSIFIABILITY ?
- needs to be able to be empirically tested to see if it false
What is QUANTITATIVE DATA ?
- numerical data
- analysed using statistical techniques
What are the STRENGTHS of QUANTITATIVE DATA ?
OBJECTIVE:
- does not require interpretation
- less prone to bias
EASY TO ANALYSE:
- computer programmes
- allows larger sample size = generalisable
What are the WEAKNESSES of QUANTITATIVE DATA ?
DOESN’T TELL US WHY:
- cause and effect
- hard to make PRACTICAL APPLICATIONS
NARROW:
- only certain behaviour can be measured this way
- reduces the SCOPE OF STUDY
What is QUALITATIVE DATA ?
- detailed information
- themes in pps responses
What are the STRENGTHS of QUALITATIVE DATA ?
RICH DETAIL:
- more representative of real life
CAN EXPLAIN WHY:
- cause and effect
- develop more accurate theories
- practical applications
What are the WEAKNESSES of QUALITATIVE DATA ?
SUBJECTIVE
- open to interpretation
- bias
- reduces VALIDITY
DIFFICULT TO ANALYSE
- transcripts have to be written
- time consuming
- smaller sample size = not generalisable
What is the MEAN ?
- average
- adding all the scores up and dividing by number of scores
What are the ADVANTAGES of the MEAN ?
- most SENSITIVE and REPRESENTATIVE
- takes all scores into account
What are the DISADVANTAGES of the MEAN ?
- distorted by EXTREME SCORES
- UNREPRESENTATIVE
What is the MEDIAN ?
- middle score
What are the ADVANTAGES of the MEDIAN ?
- unaffected by EXTREME SCORES
What are the DISADVANTAGES of the MEDIAN ?
- only looks at one or two scores
- generally used with ORDINAL DATA
What is the MODE ?
- most frequent
What are the ADVANTAGES of the MODE ?
- unaffected by EXTREME SCORES
What are the DISADVANTAGES of the MODE ?
- affected by the change in one score
- UNREPRESENTATIVE
- used with NOMINAL DATA
What are the 3 MEASURES of CENTRAL TENDENCY ?
- mean
- median
- mode
What is a MEASURE of CENTRAL TENDENCY ?
- provides a SINGLE VALUE which is REPRESENTATIVE pf a set of numbers by implicating the most TYPICAL VALUE
What are the 2 MEASURES of DISPERSION ?
- range
- standard deviation
What is the RANGE ?
- difference between the highest and lowest scores and adding 1 (allows for any rounding that has occurred)
What are the ADVANTAGES of the RANGE ?
- quick to calculate
- takes account for EXTREME VALUES
What are the DISADVANTAGES of the RANGE ?
- doesn’t provide an idea around the distribution of values around the centre
- does account for INDIVIDUAL VALUES
- affected by EXTREME SCORES
What is STANDARD DEVIATION ?
- VARIABILITY of scores from its MEAN
What are the ADVANTAGES of STANDARD DEVIATION ?
- considers ALL the scores
- sensitive
What are the DISADVANTAGES of STANDARD DEVIATION ?
- difficult to calculate
- less meaningful if data isn’t normally distributed
- distorted by EXTREME SCORES
What are BAR CHARTS ?
- vertical bars of equal distance apart
- nominal data
What is a HISTOGRAM ?
- shows distribution of scores
- continuous scale
- ordinal / interval data
What is a SCATTERGRAM ?
- plotting correlations
- visual image
What are TABLES used for ?
- descriptive statistics or percentages
What is a NORMAL DISTRIBUTION ?
- data is symmetrical
- forms bell-shaped curve on graph
What is a SKEWED DISTRIBUTION ?
- data is NOT symmetrical
What is a POSITIVE SKEW ?
- mean moves to the RIGHT
- HIGHER SCORING pps have moved mean to the right
What is an example of a POSITIVE SKEW ?
a hard test where most students didn’t score very well
What is a NEGATIVE SKEW ?
- mean moves to the LEFT
- LOWER SCORING pps have moved the mean to the left
What is an example of a NEGATIVE SKEW ?
a test which was easy and most students scored well
What is NOMINAL DATA ?
- counting frequency data
- separate categories
- each piece of data can only go into one category
What is an example of NOMINAL DATA ?
counting whether pps are happy or sad
What is ORDINAL DATA ?
- rating on a scale
What is an example of ORDINAL DATA ?
- John came first, Fred came second, Brian came third
What is INTERVAL / RATIO DATA ?
- similar to ordinal
- has a UNIT e.g. grams
What is an example of INTERVAL / RATIO DATA ?
- measure time in seconds
What are INFERENTIAL STATISTICS ?
- allow researchers to draw conclusions about their research
What is PROBABILITY ?
- psychologists need to mathematically express the likelihood their result occurred due to chance
What 3 factors determine the choice of statistical test ?
- difference or relationship ?
- experimental design type ?
- type of data ?
What is a TYPE I ERROR ?
- null hypothesis is rejected when it should have been accepted
- FALSE POSITIVE
- significant level isn’t harsh enough
What is a TYPE II ERROR ?
- null hypothesis is accepted when it should have been rejected
- FALSE NEGATIVE
- significant level is too harsh
Why do we use a 5% significant level ?
strikes a balance between making type I and type II errors
when designing a study what 3 things should be included ?
- DESIGN - experimental design, variables, controls
- MATERIALS - any special materials
- DATA ANALYSIS - reference descriptive and inferential analysis
in order to know what statistical test to use, what 3 things should you ask yourself ?
- am i looking for a DIFFERENCE or RELATIONSHIP ?
- what is my EXPERIMENTAL DESIGN ?
- what type of DATA do i have (nominal / ordinal / interval)
NOMINAL + UNRELATED DESIGN (independent groups) =
CHI-SQUARE
NOMINAL + RELATED DESIGN (repeated measure / matched pairs) =
SIGN TEST
NOMINAL + CORRELATION =
CHI-SQUARE
ORDINAL + UNRELATED DESIGN =
MANN-WHITNEY
ORDINAL + RELATED DESIGN =
WILCOX
ORDINAL + CORRELATION =
SPEARMAN’S RHO
INTERVAL + UNRELATED DESIGN =
UNRELATED t-TEST / INDEPENDENT t-TEST
INTERVAL + RELATED DESIGN =
REALTED t-TEST
INTERVAL + CORRELATION =
PEARSON’S R
How could you deal with the limitations of a repeated measure design
COUNTERBALANCING
- splits pps so they complete different levels of the IV in a different order
- balance out order effects
How could you deal with the limitations of an indepedent groups design ?
RANDOM ALLOCATION
- pps randomly allocated condition to distribute them evenly
How could you deal with the limitations of a matched pairs design ?
PILOT STUDY
- consider key variables that may effect the DV