OCR Psychology investigations revision notes (January exam)

My notes for the OCR january exam, includes; correlation, self report, experiments, observations, sampling techniques, reliabilty and validity

HideShow resource information
  • Created by: roxy
  • Created on: 09-01-12 19:44
Preview of OCR Psychology investigations revision notes (January exam)

First 292 words of the document:

OCR AS Psychology Investigations revision notes
A single number that describes the degree of relationship between two variables, particularly where it would
be difficult or unethical to experiment.
Naturalistic form of research. No manipulation of IVs or EVs.
Because it's naturalistic, can NEVER establish cause and effect
Correlational hypothesis: There will be a correlation between... (Don't forget to operationalise)
Correlational null hypothesis: There will be no correlation between....
Correlational hypothesis can be one tailed (predicts direction i.e. positive or negative) or two tailed (only
sates there will be a correlation).
Positive correlation: As one increases so does the other
Negative correlation: As one increases the other decreases
Strengths and weaknesses
Allow us to investigate topics that would normally be unethical to experiment
Provide a good starting point for wider research
Can never establish cause and effect, so they only tell us two things are related, not how and why.
Line of best fit (AKA regression line)
Has roughly equal amount of points above as below.
If it isn't obvious it is safest to say there doesn't appear to be a correlation.
Interpreting scatter graphs
If asked for two conclusions you could make from the graph...
1) General pattern correlation or not? Positive or negative?
2) Say that there is an outlier (people who don't fit the general pattern)
Self report
The participant provides information about themselves
Strengths and weaknesses
The participant knows themselves the best= high validity
It's the only way to obtain certain information (e.g. life history)

Other pages in this set

Page 2

Preview of page 2

Here's a taster:

Invalid data may be provided. This may be due to embarrassment (social desirability bias) or the
may be deceiving themselves (e.g. alcoholics)
If the question is leading (the researcher suggests an answer in the question) the data will be invalid
Structured all questions are preset
Unstructured there are no preset questions
Semistructured some questions are preset
Interviews can be part of an experiment or could be naturalistic research.
Questionnaires (AKA surveys)
Participant (self completion) or researcher completes.
Open questions the answer is not limited.…read more

Page 3

Preview of page 3

Here's a taster:

Limits the answers participants can choose from, so it is possible the answer they want to give is not
Scales can be interpreted different, practically if each value does not have a clear meaning. (e.g. the
score 3 may mean different things to different participants)
In experiments the IV's are manipulated and the EVs are standardized.
Strengths/ weaknesses of experiments
Can establish cause and effect because EVs are controlled.
They are easily replicated, which means reliability is increased.…read more

Page 4

Preview of page 4

Here's a taster:

Individual differences. It may be the participants difference that create the difference in scores, not
the IVs (so you cant conclude the IV caused the DV)
Measures of central tendency
Mean can be distorted by extraneous scores
Mode better if there are two or more `bumps' in the graph as it shows there may be more than
one most common score
Median not affected by extraneous scores
Range helps you to decide if there are ay extraneous scores
Can be naturalistic research (i.e.…read more

Page 5

Preview of page 5

Here's a taster:

Structured/ unstructured observations
Structured The researcher designs a coding scheme* to record behaviour.
Unstructured the researcher records all the behaviour they see.
*Coding schemes
A coding scheme is a list of behaviours which the observer ticks when they see that behaviour happen.…read more

Page 6

Preview of page 6

Here's a taster:

Interrater (or inter observer) reliability
Reliability= consistency!
IN observations, reliability always referred to if two observers scored independently, would they get the
same results.
Most observations have 2 observers, if not the behaviour observed could just be the opinion of the
If the 2 observers get same/ similar results, the study has high inter rater reliability
80% agreement between 2 observers is normally the target.…read more

Page 7

Preview of page 7

Here's a taster:

Stratified sampling classifying the target population into categories and then choosing a sample which
includes participants from all the categories.
Strengths and weaknesses
Strengths Weaknesses
Opportunity Quickest and easy to conduct Sample can be biased, as researcher will
choose from their social/ cultural group.
People chosen may refuse, giving an eve
more biased sample, as those who do
take part may have a certain personality.…read more

Page 8

Preview of page 8

Here's a taster:

Check your measuring system against something else which measures the same thing
Improving reliability will improve validity!…read more


Rees *Redacted*

This is some awesome, and very condensed notes :)

I'm doing a re-take, and I find these very useful.

Similar Psychology resources:

See all Psychology resources »See all resources »