Research Methods

Directional - we will state in what way one variable will effect the other. E.g. stress will increase the number of days off sick. It's specific.

Non-Directional - we are unsure of which way the variable will effect the other. E.g. there will be a difference of the number of days off between those who score higher and those who score lower on a stress test. It's not specific.

Null - when we state what will happen if there is no difference or relationship between our vairable. E.g. there is no difference between the agression scores for boys and girls.

Alternative - the same as the directional and non-directional hypothesis.

Experimental - the alternative hypothesis when used in an experiment.

Independant - the variable that the researcher manipulates and changes.

Dependant - the variable that changes as a direct result of manipulating the independant variable. It is the variable that the researcher measures.

Extraneous - variables that are needed to be controlled so they do not have an effect on the research. E.g. participant variables.

Confunding - extraneous varibales that have not been controlled. They have confuded or confused the results.

Labratory - this is when you bring the participants into a controlled environment to conduct research under controlled conditions.

+ Allows you to control extraneous variables.

- Lacks ecological validity.

- Could be considered unethical.

Field - when you go to places where the participants are. E.g. if you want to observe children's behaviour then you go to a children's park.

+ Real life setting so has ecological validity.

- Cannot control the extraneous variables.

- Considered unethical as participants do not give their consent.

Natural - this is when you use the naturally occuring situation.

+ Independant variable cannot be measure for ethical or practicle reasons.

+ Enables psychologists to measure real life situations.

- Cannot demonstrate causal relationships.

- Many extraneous variables.

Quasi - this is an experiment that you cannot control.

It is a way of establishing whether there is a relationship between two variables and assessing the strength of that relationship.

Unlike experiments, correlation studies do not tell you about causal relationships and there is no independant or dependant variables.

The correlation coefficient is a statistical method for assessing the strength of a correlation. A correlation coefficient is a number that represents the strength of the relationship between two variables.

The coefficient will lie between 1 (perfectly positive) and -1 (perfectly negative). A coefficient of 0 means there is no relationship.

+ Can be used when variables cannot be mainpulated.

+ Can rule out causal relationships.

- People often misinterpret relationship and assume cause effect.

- There may be other unknown variables that cause both variables to change.