Planning and Conducting Research Flashcards
aims, hypothesis, sampling, experimental designs, variables
research aim
statement that points out what the research aims to accomplish and desired outcomes
research question
asks what the study intends to investigate
null hypothesis
predicts there will be no difference or relationship between the variables being studied and any results are due to chance and not significant
alternative hypothesis
predicts there will be a difference or relationship between the variables being studied and the results are not due to chance and are significant
one tailed hypothesis
predicts the direction of the effect of the IV on the DV, or the direction of the correlation
two tailed hypothesis
predicts that the IV will have an effect on the DV but not in a specific direction
type 1 error
incorrectly rejecting the null hypothesis
or incorrectly accepting the alternate hypothesis
type 2 error
incorrectly rejecting the alternate hypothesis
Incorrectly accepting the null hypothesis
target population and sample
the group of people the researcher is interested in and the sample is drawn from
random sampling
- selecting participants in a way where everyone has an equal chance of being selected
- no bias in who is chosen so sample is likely to be representative
snowball sampling
- relies on initial participants recruiting additional participants
- unlikely to be representative, but easy to gather a specific sample
opportunity sampling
- selecting people who are readily and easily available
- unlikely to be representative but quick to gather participants
self selected/ volunteer sampling
- asking people to volunteer for the study
- unlikely to be representative but participants will be willing and cooperate
repeated measures design
- each participant takes part in all levels of the IV
strengths of repeated measures
- fewer participants needed
- individual differences will be controlled
- easy to compare different conditions
weaknesses of repeated measures
- participants may suffer from order effects, boredom
- lead to poorer performance on tasks
independent measures design
- participants only take part in one level of the IV
strengths of independent measures
- no order effects or boredom
- less likely to deduce the aim
weaknesses of independent measures
- participant variables - people have different abilities
- large sample needed
- time consuming
matched pairs design
Different participants used in each condition, but researcher attempts to make the two groups as similar as possible
strengths of matched pairs
- Participants only have to be tested once
- Participant variables are reduced
weaknesses of matched pairs
- time consuming
- harder to find participants that match
- large number of participants needed
behavioral categories
- objective measure used in observations
- categorizing behaviors and counting them
coding frames
- predetermined list of behaviors to attempt to cover all the behavior you will see
time sampling
- recording pre-determined behaviours at regular intervals
event sampling
- counting number of behaviours in a specified time period
likert scale
- indicates how much they agree or disagree
semantic differential scale
- choose between two extremes
- opposing descriptive words
strengths of rating scales
- easy to respond to
- quantitative data easy to analyse
- tested for reliability (test retest)
- validity can be improved b y reversing the sides to avoid bias
weaknesses of rating scales
- only quantitative data lacks detail;
- risk of response bias
- points on the scale are ordinal, can’t use stats tests
open questions
- allows participants to give a range of detailed answers
- no set response
strengths of open questions
- open questions produce qualitative data which provide greater detail
- analysis retains the detail of participants’ answers so variation in response is not lost through averaging
weaknesses of open questions
- time consuming to analyse
- analysis is subjective
- findings are individual so less generalisable
closed question
- gives participants set responses to choose from
strengths of closed questions
- easy to respond to so data is more reliable/generalisable
- easy to analyse
weaknesses of closed questions
- lacks detail
- risk of response bias
- only mode can be calculated as total is nominal data
