Research methods revision notes

used from class notes and the AQA psychology B book.

HideShow resource information
  • Created by: Lauren
  • Created on: 11-05-14 16:02
Preview of Research methods revision notes

First 282 words of the document:

Research methods- revision notes
Qualitative data- non numerical data
Thoughts and feelings
Descriptive
Real life context
Rich/ detailed data, meaningful, high validity
Low reliability, difficult to analyse
Quantitative- facts and figures
Numerical data
Specific behaviour
Involves measurements
Easy to analyse, objective, reliable
Less meaningful, less ecological validity
Pilot study- a small scale study before a full scale investigation to test whether its method appropriate,
any problem in design can be rectified.
Ecological validity- is the extent to which the task and situation is `normal' and in a real, natural setting
rather than being artificial.
Lab experiment- controlled and objective, cause and effect can be established, high reliability.
Artificial, lacks ecological validity, demand characteristics possible, informed consent not always
possible.
Field experiment- strong ecological validity, natural behaviour, removes experimenter bias.
Less control, difficult to establish cause and effect, more time consuming, not always possible to brief
Quasi- experiment- allows investigation of pre-existing variables
No random allocation, less cause and effect.
Measures of central tendency
The mean- all the scores and divide the result by the number of scores.
Median- list the scores in order of size and identify the middle one.
Mode- is found by identifying the most frequently occurring value.
The mean is preferred because it makes use of all the scores and the total. It is used in more advanced
mathematical analysis. However it is only useful if the scores are similar anomalies will give a misleading
picture of the typical value, then then mode or median is preferred.

Other pages in this set

Page 2

Preview of page 2

Here's a taster:

Measures of dispersion, Range ­ indicates the difference between the highest and lowest score. The
bigger the range, the more spread out the scores.
Standard deviation- on average, how far each score differs from the mean. Large SD indicates each
score is from the mean, small SD indicates score is very close to the mean.
Experimental methods- researcher manipulates one variable to see how this effects another variable.
Independent variables- what we change in an experiment.
Dependant variables- what we measure change in.…read more

Page 3

Preview of page 3

Here's a taster:

Methods of self-report
Questionnaires- a list of pre-set questions to which participants respond. Allow a subjective analysis.
Face to face, en masse, post, internet, phone.
Cost effective, less time consuming, easy to reach a wide range of people.
Relies on respondents returning them, difficult to generalise, response bias, people don't always give
honest responses, no researcher monitoring, misinterpretations of questions.
Interviews- researcher directly asking participants questions and record their responses.
Structured interviews- use pre-determined questions.…read more

Page 4

Preview of page 4

Here's a taster:

Representative
Time consuming
Systematic sampling- The nth member of the population is chosen.
Avoids researcher bias
Not representative
Observational studies
Laboratory observation- like a lab experiment. Controlled environment.
Offer high levels of control, cause and effect
Lack ecological validity
Naturalistic observation- observing people in their natural environment. Unconstrained behaviour.
High levels of ecological validity, findings should be able to generalise to real life.
Many uncontrolled variables, cause and effect can't be generalised.
Covert observations- psychologists observe without people being aware of it.…read more

Page 5

Preview of page 5

Here's a taster:

Zero correlation- no clear relationship between the two variables. Co-efficient of 0.
Can establish the strength and direction of the relationship between variables.
Statically analyse naturally occurring phenomenon.
Cannot reliably establish a cause and effect.
Lack construct validity.
Case studies
An in-depth investigation of one person of group or organisation.
Unstructured interviews, observations, past records, medical histories, school files.
Often used to investigate atypical behaviour or unusual situations.
Sometimes used because a psychologists wants to have more insight into a particular individuals.…read more

Page 6

Preview of page 6

Here's a taster:

Ethics
Respect- value the dignity and worth of all individuals
Avoid prejudice
Confidentiality
Briefed
Audio, video, recording- consent
Informed consent
Under 16- parental consent
Not pressurised
Avoid deception
Right to withdraw
Competence- psychologists should work within the limits of their knowledge, skill, training, education
and experience.…read more

Page 7

Preview of page 7

Here's a taster:

Presenting data
Bar graphs- when data is divided into categories/ discrete data.
Line graphs- data must be numerical. Continuous data. Can be used to make estimates.
Histograms- represents continuous data. Bars represent each score- drawn touching to show the data is
continuous. Useful when there is a large scale of data.
It's easier to visually represent and compare two or more sets of data.
Scatter grams- used to represent relationships between variables- correlation.…read more

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all resources »