Research Methods Flashcards
What is random sampling? + Evaluation
Participants are picked out of a target population using either a number generator or name-out-of-hat method.
Evaluation:
- Free from researcher bias
- Time-consuming
- Representive sample is not guaranteed (some sub-groups may be overrepresented or not selected)
What is stratified sampling? + Evaluation
Target population divided into strata (groups). From each strata participants are randomly selected.
Evaluation:
- Avoids researcher bias
- May lack 100% representation
What is systematic sampling? + Evaluation
Pick every nth participant from target population.
Evaluation:
- Free from researcher bias
- Creats a fairly representive sample
What is volunteer sampling? + Evaluation
This is when the participants put themselves forward i.e. they self-select. The researcher places an advert in a newspaper or magazine or on a public noticeboard.
Evalutation:
- Requires minimal input from researchers
- Introduces demand characteristics
- Participants will be engaged
- Volunteer bias
What is an opportunity sample? + Evaluation
Opportunity sampling is where a researcher selects participants based on their availability. One example would be standing on the street asking passers by to join the research.
Evaluation:
- Convenient
- Time-effective
- Less costly
- Unrepresentative of the target population
- Suffers from researcher bias
What are lab experiments?
Lab experiments take place in controlled environments.
What is a structured interview?
Where there is a set list of questions that is not deviated from.
What is a semi-structured interview?
Where there is a list of questions but the interviewer is free to add supplementary questions, leave questions out or follow interesting lines of inquiry.
What is an unstructured interview?
There are no pre-set questions, just a free-flowing conversation. The researcher would have some themes or headings to work with, but otherwise, the discussion could go anywhere.
What are the 4 types of observations?
- Participant (the researcher plays an active part)
- Non-participant (the researcher simply observes)
- Overt (everyone is aware the observation is taking place)
- Covert (the people being observed are not aware the observation is taking place)
Define directional hypotheses and when they would be used?
A directional hypothesis is a hypothesis that predicts an outcome or direction of the relationship between variables.
It would be used when there is previous research available which informs the researcher of the expected direction of the result.
Define non-directional hypotheses and when they would be used
A non-directional hypothesis is a hypothesis that does not state the direction of the outcome.
It is used when there is no prior research in the topic or present research is contradictory. It states a relationship but not the expected direction of the results.
Define operational hypotheses
Operational hypotheses predict exactly the outcome.
What is the mean and how do you work it out?
The mean is a measure of central tendency in statistics. It gives you an average value that represents the “typical” value in the data set.
To calculate the mean, you add up all the values in a set of data and then divide the sum by the total number of values.
The mean may not be the best choice sometimes as it is affected by outliers in the data set.
What is the median and how do you work it out?
The median is another measure of central tendency. To find the median, you arrange the values in a data set in ascending or descending order and then find the middle value. The median represents the value that separates the higher and lower halves of the data. If there is an even number of values, you take the average of the two middle values. For example:
2, 4, 6, 8, 10, 12, the middle two values would be 6 and 8. Therefore, the average would be 7.
What is the independent variable?
What you change
What is the dependent variable?
What you measure/observe
What is the control variable?
What you keep the same
what are the 3 experimental designs?
- independent groups
- repeated measures
- matched pairs
describe independent groups
2 separate groups experience 2 different conditions
describe repeated measures
all participants experience both conditions
describe matched pairs
participants paired together on a variable/variables relevant to the experiment
what are strengths of independent groups?
- order effects aren’t a problem
- participants less likely to guess aim
what are limitations of independent groups?
- independent differences acting as a confounding variable meaning validity is reduced ➜ random allocation fixes this
- less economical as each participant contributes a single result meaning increased time/money on recruiting participants
what are strengths of repeated measures?
- participant variables are controlled ➜ higher validity
- fewer participants are needed ➜ save time/money
what are limitations of repeated measures?
- order effects ➜ fixed by counterbalancing
- repeating 2 tasks can lead to boredom + fatigue so deterioration in second condition
- on skill based tasks participants may perform better in second condition due to practice
- more likely participants will figure out aim of the study ➜ demand characteristics
what are strengths of matched pairs?
- participants take part in a single condition so order effects + demand characteristics are less of a problem
what are limitations of matched pairs?
- participants can never be matched exactly ➜ important differences between participants that may affect DV
- time consuming + expensive
what is a field experiment?
IV manipulated in a natural more everyday setting which is common to the participants
what is a natural experiment?
similar to lab/field but researcher has no control over IV and cannot change it
what is a quasi-experiment?
IV that is based on existing differences between people e.g. age, gender. Unlike the natural experiment IV cannot be changed by ANYONE. DV may be naturally occuring.
what are strengths of laboratory experiments?
- high internal validity ➜ variables very controlled , easily distinguish cause + effect
- reliability ➜ more controlled means it easier to replicate than other types of experiments.
what are limitations of laboratory experiments?
- lack generalisability➜ low external validity ➜ artificial environment so behaviour may have been altered meaning participants would act differently in an everyday setting
- demand characteristics
- low mundane realism ➜ artificial tasks
what are strengths of field experiments?
- higher mundane realism ➜ natural environment (higher generalisabilty + ecological validity)
- more valid behaviour especially if covert (high external validity)
what are limitations of field experiments?
- cause and effect harder to distinguish
- ethical issues if covert
- less control over variables ➜ harder to replicate
what are strengths of natural experiments?
- typically have high external validity because of studying real life issues and problems e.g. Romanian orphan studies
what are limitations of natural experiments?
- naturally occurring events may only happen rarely reducing oppurtunities for research.
- research may be conducted in a lab ➜ lack ecological validity ➜ demand characteristics
- hard to clearly see cause and effect (only in independent group design)
- cannot randomly allocate participants to conditions ➜ confounding variables
what are strengths of quasi-experiments?
- controlled conditions ➜ variables controlled ➜ repeatability ➜ reliability ➜ internal validity
what are limitations of quasi-experiments?
- cannot randomly allocate participants to conditions ➜ confounding variables
- IV not deliberately changed therefore cannot claim IV caused any observable change
what are the 5 types of sampling?
- random
- systematic
- stratified
- opportunity
- volunteer
what is random sampling?
all members of target population have an equal chance of being selected.
selected through lottery method (computer/phone randomiser or name-out-of-hat technique)
what is systematic sampling?
every nth member of target population is selected. A sampling frame is produced (list of target population organised e.g. alphabetical). a sampling system is nominated (e.g. 3rd, 6th or 8th)
what is stratified sampling?
- identify the different strata within target population
- proportions needed for sample to be representative is worked out
- participants selected randomly from each strata
what is an opportunity sample?
select anyone willing and available
what is a volunteer sample?
participants selecting themselves ➜self-selection
researcher may put out an advert or poster etc for participants to select themselves
what are strengths of using random sample?
- unbiased ➜ confounding and extraneous variables should be equally divided amongst different groups ➜ increased internal validity
what are limitations of using random sample?
- time consuming + difficult ➜ finding full list of target population
- may be unrepresentative
- selected participants may refuse to participate
what are strengths of using systematic sample?
- reduced researcher bias
what are limitations of using systematic sample?
- time consuming
- participants may refuse to participate
what are strengths of using stratified sample?
- representative ➜ increased genralisability to the wider population
what are limitations of using stratified sample?
- identified strata cannot all the ways that people are different ➜ complete representation of wider population is impossible
what are strengths of using opportunity sample?
- convenient ➜ less time + money spent (no need for list of target population etc)
what are limitations of using opportunity sample?
- unrepresentative ➜ drawn from specific area ➜ low generalisability
- researcher bias
what are limitations of using volunteer sample?
- gets a certain profile of person that is more curious and more likely to try and please the researcher (demand characteristics) ➜ negatively impacts generalasiability
what are strengths of using volunteer sample?
- convenient ➜ dont have to spend money or time
- get more engaged participants
what are the 4 ethical issues?
- informed consent
- deception
- protection from harm
- privacy and confidentiality
how do you deal with informed consent?
- participants should be issued a consent letter/form detailing all relevant information which could affect their choice of participating which needs to be signed.
other ways to get consent ➜ presumptive consent, prior general consent, retrospective consent
what are the 3 other ways to get consent that isnt a form/letter?
- presumptive consent ➜ similar group of people are asked if study is acceptable if this group agrees consent of participants is presumed
- prior general consent ➜ participants give their permission to participate in a number of different studies (including one with deception)
- retrospective consent ➜ asked for consent during debrief (may not of known they were being studied or subject to deception)
how do you deal with deception and protection from harm?
- debrief detailing true aims of study + any details that were not supplied during the study
- told what their data will be used for and given right to withdraw during study and right to withhold data
- counselling should be provided by researcher if stress etc affected participant
how to deal with confidentiality?
- personal details protected
- typical that there is no persona data of participants instead they are referred to by initials of numbers
- stated that during briefing and debriefing that data will be protected + data won’t be shared with other researchers
what is a single-blind procedure?
participants not knowing about the true aim of the study
what is a double-blind procedure?
participants and researchers aren’t aware of true aims of the study (only researcher who is conducting it knows the true aim)
