Correlations
Correlations are a way of establishing whether there is a relationship between two variables. Correlations can be positive – where one variable increases so does the other; or negative – as one variable increases, the other decreases.
Correlations do not identify the cause, they just indicate whether there is a relationship or not – both variables are equal (no IV or DV)
Correlation coefficients are a statistical method for assessing the strength of a correlation. The sign (+ or -) tells you the direction of the correlation. The number (between 0 and 1) tells you the strength
+ Allows study of hypothesis which cannot be studied directly – it is useful for identifying relationships between variables that might not have been considered previously
+ Fairly straightforward to conduct – few controls needed
+ Identifies further areas for research – shows relationships that may not have been expected
- Interpretation of results difficult.
-Cause and effect not established.
- Direction of cause is uncertain.
- Other variables may be acting.
Case Studies
Case studies are a research method that allows data to be collected in depth and detail
It can be the study of one individual or a small group of individuals
Some case studies can be funded e.g. government studies when focusing on a particular issue
Case studies use interviews, questionnaires and observations. This data is then put into a case history which uses qualitative data. Using different techniques to gather data is called triangulation
Qualitative data is analysed by looking at themes and cross case analysis
+ A lot of detailed information is gathered
+ Gives a complete picture to reasons why and focuses on real life
+ Valid data is produced – natural setting so ecologically valid, has construct validity – measures what it has claimed and is true to life
- Only one person or a small group is studied – focuses on the individual and can therefore be unique however lacks generalisability
- Less reliable as unable to repeat them – difficult to assess whether the information it produces is useful
- Unscientific
Freud’s case study methods: Free Association, Dream Analysis, Slips of the tongue. Freud used case studies and these methods to access the unconscious
+ Qualitative data – lots of detail
+ Valid – from real situations difficult to gather data using other methods
+ Freud’s methods can be used as therapy as well as a research method - Method of understanding the unconscious and retrieving repressed memories (therapeutic)
- Free association
- Interpretation of data
- Accessing the unconscious and this is not measurable in a scientific way
Longitudinal Studies
Longitudinal studies are used to study the same participants over a set period of time e.g. BBC Child of Our Time is following families from pregnancy until the children are 20
+ useful for looking at developmental trends e.g. how development affects different characteristics, able to establish norms e.g. for walking
+ Uses the same participants
- Participants may drop out – may lead to a biased sample
- Researchers may change, move away or lose funding for the study
Cross Sectional Studies
Cross sectional designs use a measure taken at one moment in time and results measured and compared. E.g. language acquisition – sample 2 and 5 year olds and compare the data
+ Immediate results, cheaper, easier to organise
+ More ethical
- Different participants – individual differences e.g. upbringing
- Other variables may not be controlled and may affect findings
Types of data:
Nominal – categories are recorded e.g. number of yes/no responses (numbers are numbers in categories and not scores)
Ordinal – ranking data, number applies to scores e.g. rank attractiveness on a scale of 1 – 5
Interval/Ratio – numbers have equal intervals and can be treated mathematically e.g. height, weight
Correlations are a way of establishing whether there is a relationship between two variables. Correlations can be positive – where one variable increases so does the other; or negative – as one variable increases, the other decreases.
Correlations do not identify the cause, they just indicate whether there is a relationship or not – both variables are equal (no IV or DV)
Correlation coefficients are a statistical method for assessing the strength of a correlation. The sign (+ or -) tells you the direction of the correlation. The number (between 0 and 1) tells you the strength
+ Allows study of hypothesis which cannot be studied directly – it is useful for identifying relationships between variables that might not have been considered previously
+ Fairly straightforward to conduct – few controls needed
+ Identifies further areas for research – shows relationships that may not have been expected
- Interpretation of results difficult.
-Cause and effect not established.
- Direction of cause is uncertain.
- Other variables may be acting.
Case Studies
Case studies are a research method that allows data to be collected in depth and detail
It can be the study of one individual or a small group of individuals
Some case studies can be funded e.g. government studies when focusing on a particular issue
Case studies use interviews, questionnaires and observations. This data is then put into a case history which uses qualitative data. Using different techniques to gather data is called triangulation
Qualitative data is analysed by looking at themes and cross case analysis
+ A lot of detailed information is gathered
+ Gives a complete picture to reasons why and focuses on real life
+ Valid data is produced – natural setting so ecologically valid, has construct validity – measures what it has claimed and is true to life
- Only one person or a small group is studied – focuses on the individual and can therefore be unique however lacks generalisability
- Less reliable as unable to repeat them – difficult to assess whether the information it produces is useful
- Unscientific
Freud’s case study methods: Free Association, Dream Analysis, Slips of the tongue. Freud used case studies and these methods to access the unconscious
+ Qualitative data – lots of detail
+ Valid – from real situations difficult to gather data using other methods
+ Freud’s methods can be used as therapy as well as a research method - Method of understanding the unconscious and retrieving repressed memories (therapeutic)
- Free association
- Interpretation of data
- Accessing the unconscious and this is not measurable in a scientific way
Longitudinal Studies
Longitudinal studies are used to study the same participants over a set period of time e.g. BBC Child of Our Time is following families from pregnancy until the children are 20
+ useful for looking at developmental trends e.g. how development affects different characteristics, able to establish norms e.g. for walking
+ Uses the same participants
- Participants may drop out – may lead to a biased sample
- Researchers may change, move away or lose funding for the study
Cross Sectional Studies
Cross sectional designs use a measure taken at one moment in time and results measured and compared. E.g. language acquisition – sample 2 and 5 year olds and compare the data
+ Immediate results, cheaper, easier to organise
+ More ethical
- Different participants – individual differences e.g. upbringing
- Other variables may not be controlled and may affect findings
Types of data:
Nominal – categories are recorded e.g. number of yes/no responses (numbers are numbers in categories and not scores)
Ordinal – ranking data, number applies to scores e.g. rank attractiveness on a scale of 1 – 5
Interval/Ratio – numbers have equal intervals and can be treated mathematically e.g. height, weight
