Research Methods Flashcards
whats an aim?
a general statement of what the researchers intend to find out in a research study
whats a hypothesis?
a precise testable statement about the assumed relationship between variables
whats a directional hypothesis?
states there will be a change and the direction in which it will go in
whats a non directional hypothesis?
predicts a change but not the direction of the change
whats a null hypothesis?
no significant difference exists
whats the independent variable?
characteristic that changes
whats the dependant variable?
being measured
whats operationalisation?
clearly defining the variables in terms of how they can be measured
what do psychologists do in a study? (the order) four steps
theory - prediction - experimental - conclusion
“lack of sleep affects reaction time” is this directional or non?
NON
“”high temperatures grow tomatoes quickly” non or directional?
directional but bad
what is the IV and DV of “sleep affects how quickly you complete a crossword”?
IV- less than or more than 8 hours of sleep
DV- time how many seconds it takes them to complete it
define extraneous variables
any variable luch affects the DV if not controlled
state what four things extraneous variables are divided into and define them?
• demand characteristics - participants know the aim and therefore change their behaviour
• investigator effects - any effects of the researchers behaviour on the outcomes of the research/DV
• participant variables - individual differences between participants that may affect the DV
• situational variables - any features of the experimental situation that may affect the DV
give examples of:
• investigator effects
• participant variables
• situational variables
from the two vs no sweet memory experiment in class
• sampling tech used, how they interact with participants
• IQ, amount of sleep
• noise, view of the board
define standardisation?
• using exactly the same procedures and instructions for all participants
example of standardisation?
• same words, same time, same room/ situation
define randomisation
• the use of chance to control bias
eg. names in hat at 1 to 2 method
what is the aim of any psychological study?
• provide information about how people behave in ‘real life’- the everyday settings which life is lived
what is the risk on ppt if the study is too artificial
they will not act as they normally would
name and define the different experimental design?
• repeated measures design - there is only one group of participants that take part in both conditions
• independent groups design - there are two separate groups of participants (one for condition A one for B)
• matched pairs design - two separate groups matches into pairs for certain qualities eg age
evaluation of repeated measures design and how to deal with limitations?
pos: removed participant characteristics
neg: order effects, increase demand characteristics, conditions may differ
dealing with it: counterbalancing
evaluation of independent groups design and how to deal with limitations?
pos: order effects removed, less chance of demand characteristics
neg: more ppt needed, lots of participant variables uncontrolled, need to be careful with standardisation
fix it: make groups as similar as pos, randomly allocate to remove bias
evaluation of matched pairs design and how to deal with limitations?
pos: decrease ppt variables, order effects not an issue
neg: difficult to match up
fix it: conduct a pilot study (small scale version) to consider key variables to match
state where it occurs and describe:
- order effects
- counterbalancing
- random allocation
- repeated measures = practice (knowing what to do, bias), fatigue (tired so does worse)
- repeated measures = “ABBA” half of the participants do A first then B others do B then A
- independent groups = they are allocated to a condition (A or B) by tossing a coin or picking a number
define reliability?
(being repeatable)
• different researchers doing the same method gets the same results - consistency, high reliability, standardisation, high control
define validity
(is it truthful)
• does it measure what it was supposed to measure
• the extent to which an observed effect is genuine
define internal validity
whether the effects of an experiment is due to the IV and not EVs
• low internal val= high amount of EVs
define external validity
(mundane realism)
(ecological validity)
whether it can be generalised to the outside world
low = too artificial
name the four types of experiments:
laboratory
field
natural
quasi
describe a laboratory experiment
• take place in a controlled artificial environment - usually in uni’s
• EV’s controlled - clear cause + effects established and very controlled
• standardised procedures
• most scientific form
• objective - free from bias - researcher detached
describe a field experiment
• experiment in everyday real life setting
• IV -> DV
• extraneous variables different to control
• particpants are unaware - demand characteristics lowered
explain a natural experiment
• researcher doesnt manipulate the IV, occurred naturally
• used when its not practical or ethical to manipulate IV
define a quasi experiment
is conducted when the IV is based an existing differences between
people
name the strengths and weaknesses of a labatory experiment?
strengths:
• high reliablity
• free from bias
• very controlled
• high internal validity
• EV’s controlled
weaknesses:
• overartificial - lacks realism/ external validity
• investigator effects can occur
• internal validity could be compromised - demand characteristics
name the strengths and weaknesses of a field experiment?
strengths:
• the participants are unaware - removes chance of demand characterstics.
• higher mundane realism / external validity
• more research opportunities
weaknesses:
• lower internal validity - more EVs.
• low reliability
• ethical issues - not consensual.
name the strengths and weaknesses of a natural experiment?
strengths:
• high realism/ external validity
• research opportunities in areas that wouldnt be possible otherwise.
weaknesses:
• low internal validity - unknown EVs
• low reliability
• lack of opportunity
• demand characteristics - bias
define an ethical issue?
• these arise when a conflict exists between the rights of the ppts in research studies and the goal of research to produce authentic, valid and worthwhile results
name the 5 guidelines for the British Psychology Society?
• consent (informed)
• deception
• confidentiality and privacy
• right to withdraw
• protection from harm
explain informed consent
ppt should know as much as possible ab the procedure before they agree to do it (eg, how long it will take, what they will be asked to do)
• they should also be aware of the aims and their rights to be able to withdraw at any point
• some believe that informed consent makes a study meaningless because ppts do not act naturally if they know the aims
explain deception?
deliberately misleading our participants in some way - debrief after !!!
explain confidentiality and privacy?
• legal rights for ppts, researchers should do everything possible to ensure anonymity by removing names and identifying details
• ppt have the right to control info ab themselves
explain protection from
harm?
• PPT SHOULD BE LEFT IN THE SAME WAY THEY ENTERED
• psychological and physical harm
• allowed to withdraw if uncomfortable
• should NOT be subject to embarrassment, stress or under pressure
whats the mnemonic to remember ethical guidelines?
Can - confidentiality
Do - deception
Cant - consent
Do - DEBRIEF (after)
With - withdraw
Participants - protection of ppts
what is the target population?
whole group that you are studying
whats the sample frame
a list of everyone in the population
whats the sample
group of people taken from sampling frame on who you do the research
what is cross sectional
it will be made up of a range of different people to best represent the research population
whats generalisation
research from the study can be applied to the whole research population
whats representative
when data can be said to accurately represent the research population in terms of eg gender and age
name the 5 different sampling techniques ?
random, opportunity, volunteer, stratified/quota, systematic
random sampling…
• names out of a hat or number generator
• chance NOT bias
• every member of the population has an equal chance of being chosen
• not representative for small sample but can be for large
• time consuming
opportunity sample…
• economical
• whenever is available at the time and willing to be studied
• bias - ask certain ppl
• might not be representative
volunteer sampling…
• people actively volunteer to be in a study by responding to a request which has been advertised
• consent is more ethical
• bias - only volunteer if your interested
stratified/ quota sampling…
• mathematically worked out - reflect population
• time consuming
• representative
• classify population into categories and choose sample which consists of ppt from each category to reflect population
systematic sampling…
• bit random
• list and take nth number
• researcher picks ppt at set intervals from sampling frame
• not bias
• not necessarily representative especially with small groups
what are the ppts and researchers POV for informed consent?
ppt: necessary when deciding whether to take part
researcher: may cause demand characteristics
what are the ppts and researchers POV for deception?
ppt: prevents ability to give informed consent
r: may cause demand characteristics
what are the ppts and researchers POV for right to withdraw?
ppt: may feel unable to as it would spoil the experiment, some studies offer rewards so they nay feel unable to leave
r: loss of ppts = bias findings
what are the ppts and researchers POV for protection from harm
ppt: no desire to be harmed
r: may be impossible to predict risks
what are the ppts and researchers POV for confidentiality?
ppt: data protection act makes confidentiality a legal right
r: details of a study may lead to ppt identification
what are the ppts and researchers POV for privacy?
ppt: people have an expectation of privacy
r: demand characteristics
how could each issue be dealt with?
- informed consent
- deception
- right to withdraw
- protection from harm
- confidentiality
- privacy
- formally agree (in writing) and debrief after the study and offer right to withhold data
- seek permission from an ethical committee, debrief
- emphasis at the start, remind that throughout study assured of reward even if they leave
- avoid risks greater than those in everyday life, stop immediately
- no names (numbers instead), store information securely
- do not observe without consent, if not usual observing situation, maintain confidentiality
what are pilot studies?
small scale version of an investigation that takes place before the real investigation is conducted
define single blind trial?
• ppts unaware- researcher aware
• any info that might create expectations is not revealed until end of the study -> avoid demand characteristics
define a double blind trial
neither the ppt or researcher who conducts the study is aware of the aim
• rule out placebo - cant influence behaviour
define control group in trials
setting a baseline
• neutral group to formulate comparisons
what is a questionnaire?
• a set of questions designed to collect info about a topic/s
• strengths =
- collect the same info from a large n of people easily
- can access what people think (observations etc rely on guessing what people think)
describe open question?
ppt can give any answer they wish
define closed questions
set number of responses that ppt have to select from
define a likert scale
a scale of responses to a question often 1-5 demonstrating a level of agreement
define a rating scale
assesses the strength of a ppt ls opinion again - ppts must identify a value that represents their strength of a feeling
what type of data (qualitative or quantitative) is these for:
• open question
• closed question
• likert scale
• rating scale
• quantitative
• qualitative
• quantitative
• quantitative
what is a structured interview?
• interview questionnaires
• list of q the researcher reads out in a particular order
• closed questions
• largely quantitive data
whats an unstructured interview?
• relaxed environment - at ease
• not preplanned q’s
• less formal
• more like a convo
whats a semi structured interview?
• combines both structured and unstructured interviews
• The interviewer might have a series of prepared questions, but will alter the interview as it goes along, adding new questions as information arise
• It’s rather like the interview you have when you go see a doctor, and as such, is often called a clinical interview.
advantages and disadvantages of open question questionnaires?
adv:
- higher validity
- can answer how they want = new insights
disadv:
- difficult to analyse
- low response rate
- not flexible
advantages and disadvantages of closed question questionnaires?
adv:
- quantitive data, easy to analyse
- easier to obtain a large sample
- easy to replicate, high reliability
dis:
- low validity, not a prefixed response which represents the responded feeling
advantages and disadvantages of a structured interview?
adv:
- easy to repeat
- less interview skill needed
- quantitative data
- easier to analyse
dis:
- sample size is smaller than questionnaire
advantages and disadvantages of unstructured interviews?
adv:
- qualitative data
- detailed info obtained
disadv:
- sample size smaller than questionnaire
- skilled interviewer
- interview bias
what is an observational technique
observational studies involve watching and recording peoples behaviour
what does non experimental methods mean?
eg observations
• no manipulation of variables
• not classed as experiments
• researcher observe behaviour and look for patterns
• cant establish cause and effect relationships
• observations are used in psych research in one of two ways, a non experimental method or as part of another research method
what is a correlation?
A correlation is a relationship between two variables. A correlational design is a way to test the relationship between two variables. A participant provides data for both variables. A correlational design allows us to test whether 2 or more phenomena are related, and if so how strongly.
whats in a correlation hypothesis?
states the expected relationship between co variables
advantage of correlation analysis?
a good preliminary method for future research
limitation of correlation analysis
?
no clear cause + effect
what is quantitative data?
- data that focuses on numbers - numerical data
what is qualitative data?
- data that describes meaning and experiences expressed in words
what type of data does experiments produce?
quantitative - DV is measured to establish cause and effect
what type of data does observations produce?
- unstructered = qualititive
- coding system + behavioural catergories = quantitative
what type of data does self report techniques produce?
- questionnaires =
• likert, closed, rating are quantitive
• open questions are qualititive - interviews =
• structered are N
• unstructured are L
what type of data does correlations produce?
quantitative
evaluation of the types of data?
• quantitative -
easy to analyse
oversimplifying behaviour
• qualitative -
detailed explanation
detailed insight so hard to analyse
whats primary data?
information that has been obtained first hand by the researcher
whats secondary data?
information that has already been collected by someone else and so pre dated the current research
positives of primary and secondary data
primary = new and relevant
secondary = easy and free
negatives of primary and secondary research?
primary = time and effort
secondary = could be out of date
what does a normal distribution sketch look like?
bell curve
identical
features of normal distribution
mean mode and median at all same point
data symmetrically spread
dispersion of scored either side of middle is consistant
what does a negatively skewed distribution look like and features?
mean is LOWER than mode and median
LEFT FOOT !!!!!
what does a positively skewed distribution look like and features?
RIGHT FOOT !!!!
mean is higher than rest
standard deviation is?
tells us how varied ppt scores are
what does a small standard deviation mean?
ppts are performed similarly
what does a large standard deviation mean?
lots of variation between ppts
in the sign test what does each thing mean?
P
N
SV
CV
P - probability MUST BE LESS THAN OR EQUAL TO 0.05
N - number of ppts minus the zero scores
S/SV - how many + or - and whichever has the lowest freq
CV - on the table, one or two tailed, N and across
does the S value need to be less or more than the CV foe the hypothesis to be accepted and results to be significant
less than or equal to
three reasons to use sign test
- repeated measures
- test of dif
- nominal data
what does the sign test do?
determine whether the difference we have found in our research is significant
what is peer review ?
the assessment of scientific work by others who are specialists in the same field, to ensure any research intended for publication is of high quality
what is the purpose and aim of peer review?
the report is considered in terms of its validity (is it truthful?), significance (importance?) and originality (is it unique?)
