Developing a study
An aim is on sentence that clearly highlights what the researcher is intending to investigate.
A hypothesis is a testable statement about the relationship between two variables. The two types of variables are independent and dependent. A hypothesis must include how the researcher predicts that the IV will affect the DV.
The independent variable (IV) is the variable that's manipulated by the person carrying out the experiment.
The dependent variable (DV) is the variable that's measured to see if the IV has affected it.
There are three main types of experimental environment:
- Lab experiment
- This is when the experiment is carried out in tightly controlled conditions.
- An advantage is that they allow extraneous variables to be controlled, so accurate conclusions about cause and effect can be made. Another advantage is that they can be repeated so that the results can be checked.
- A disadvantage of this type of experiment is that it may lack ecological validity because it doesn't reflect on normal behaviour. Another is that the participants may behave differently, which could make the results misleading.
- Field experiment
- When an experiment is carried out in everyday situations, but still with the control of the IV.
- An advantage of this is that the behaviour of participants is more natural, so it has greater ecological validity.
- A disadvantage is that not all of the variables are controlled, so another factor may have influenced the results.
- Natural experiment
- When a study is conducted in a situation that cannot be manipulated. The IV occurs naturally.
- Have very high ecological validity because the behaviour is completely natural.
- Conclusions cannot be drawn about cause and effect because there are too many uncontrolled variables.
Pros and cons of experiments in general
- Can control variables so we can identify cause and effect.
- The experimenter can manke sure that no other factors are effecting the dependent variable.
- Experiments can be repeated by other researchers to because they're controlled and standardised.
- They're objective because experimenter bias is controlled.
- They're often carried out in artificial settings that are unrealistic to everyday life. Therefore they lack ecological validity.
- Demand characteristics could become present as they participants try to chage their behaviour.
- Repeated measures
- When one group of participants take part in all conditions of the experiment.
- Advantages are: The researcher can make a direct comparison between the results. There are no participant variables.
- Disadvantages are: Participants could guess the aim of the study because they take part in all conditions. They may get better as the experiment goes on (order effect).
- Matched pairs
- When the groups of participants are matched to similar skills, abilities, and qualities. Each group takes part in one condition of the experiment.
- Advantages are: The participants are matched on key variables, so the participant variables are reduced, which would allow the researcher to gather more accurate results. There is no order effect.
- Disadvantages are: Matching participants is difficult and can take a long time. Some variables are still present.
- Independent groups
- When participants only take part in one condition of the experiment.
- Advantages are: No order effect. The same tasks and materials can be used for each condition.
- Disadvantages are: There are a lot of participant variables, which will effect the results. You'll need more people to take part in the study.
These are the right standards of behaviour that we should use towards other people. Experiments consider:
- Deception - P's should not be lied to about the experiment, and if deception is used the p's should be debriefed to explain why.
- Consent - P's should be told the aim of the study and what it will entail, and be able to choose whether they take part or not.
- Debriefing - P's should be fully informed about the study at the end of the experiment..
- Right to withdraw - P's should have the right to remove themselves and their data from a study at any time, and should be reminded of this.
- Confidentiality - P's details should be secure and private. If their data is published, the participants must give permission first.
- Use of children - Parental consent must be given and the study should be stopped if the child becomes distressed.
- Protection - P's should be protected form harm during a study, and shouldn't be put under stress.
Samples are representative of the target population that is being studied.
- Systematic - When participants are selected at fixed intervals from a list of the target population.
- Pros - Quick and easy to do, and eliminates researcher bias.
- Cons - If the sample isn't randomised, a hidden order may appear (making the sample unrepresentative).
- Stratified - When sub-categories are identified in the target population and the participants are selected in proportion to these.
- Pros - v representative of the target population. Unbiased.
- Cons - Takes a long time and can be expensive.
- Opportunity - When people that are ready and willing are selected to take part in an experiment.
- Pros - Quick and easy way to gather participants.
- Cons - Sample may not be representative. Could be researcher bias.
- Random - When every member of the target pop. has a fair chance of being represented.
- Pros - Quick, easy and cheap.
- Cons - Can be time consuming and difficult
Standardised procedures are when a researcher makes a set order of carrying out a study, thats applied to all of the participants taking part. They're used to minimalise the effect of extraneous variables. These are variables over than the IV that could affect the DV if they're not controlled.
Randomisation is used to ensure that procedures are unbiased. Random allocation is when participants are put into groups by chance.
To make sure that all participants have the same experience, they do the experiment in the same place, in the same conditions (temp. etc), and are all delivered the same instructions in the same way (Eg. on paper).
Counterbalanicing is used to "even out" the order that participants take part in both conditions of an experiment.
Mean - the sum of all the scores divided by the number of scores.
Median - the middle value.
Mode - the most frequent value.
Range - the largest score minus the smallest score.
An anomalous score is one that doesn't fit the trends of the data (outliers). They're too high/low in comparison to the other results.
Raw data is data that has not yet been analysed.
Tables are used to summarise data, bar charts are used to summarise data that is in separate groups, and line graphs are used to plot continuous data when you want to show a trend.