Contents, so to speak..
Want to know what I'll be looking at with you? Well, These are a list of the different types of Research Methods I will be focusing on. Lab, Field, and Natural are the most common (and arguably, important, so make sure you understand those. If not, ask your teacher, as I can't claim to know everything, as Much as I'd like.)
- Case Study
Lab Experiments are..
- Done in controlled, artificial environments (A Laboratory, obviously..)
- Able to completely controll all the variables.
- Highly objective and systematic. With a completely controlled, standardised procedure.
- Easy to obtain large amounts of detailed information
- Allows for easy replication - Repeated tests -> Reliability!
- Artificial environment may lead to demand characteristics (guessing purpose of experiment and reacting to either fit to/go against expectations.) or unnatural reactions/behaviour, - Thus can't be applied to real life?
- Ethical concerns - eg. Applying unnatural levels of stress, etc.
- There may experimenter bias. - eg. Bribery to make participants react in a certain way, having favourites..
Field experiments are...
- In a natural setting- ie. a street, school, supermarket.
- Controlled, much like a Lab. Experiment, only in a more aplicable to real life setting.
- High External Validity (Meaning it is more generalisable, it's similar to real life situations.)
- High degree of controll, as with Lab. Experiments.
- Little or no Demand characteristics, as Participants are reacting naturally, as they are unaware it is an experiment.
- Ethical concerns, such as lack of consent.
- Extraneous variables (Interfences from outside world, like loud noises, etc.)
- Difficulty in replicating, as exact situation may change (eg. Participants & their moods)
- Sample bias: Experimenter may only pick certain people, or for example, mothers are more likely to be found in supermarkets, or Teachers at schools.
Natural Experiments are ...
- Any situation in which experimenter has no controll over variables (eg. British people's patriotic mood around the Jubilee - You have to wait for the jubilee, you can't change the day!)
- Quasi-Experiments: The independent variable is not manipulated by the researcher.
- Participants aren't aware they are being tested on.
- (See Field experiment, as they are extremely similar!)
- Have to wait for a certain event, making repition difficult, ( and who likes waiting?! )
- Extreneous variables
- (Again, See Field Experiment, but the Ads&Disads listed here are the main few, I would say! But learn them all!)
Case Studies are...
- A detailed study or evaluation of a single indivual, event or group.
- Produces rich, detailed data on the specific thing you are studying
- Cannot generalise the findings, as they can only be applied to that individual study
- Findings are from one indivudual, and thus are not easily repeated.
- Questions, like a written form of interview.
- Gain large amounts of data quickly and cheaply. (What do you mean 'cheaply' isn't a a good word!?)
- Only the people willing to participate in the questionaire will complete it, creating bias on people on have the time, or are interested in the study will take part.
- Questionaires take skill, you have to be able to read/write, so may get some bias on that everybody who enters will be educated, or not a child.
- People can lie, meaning demand characteristics.
Correlational Experiments are...
- Testing a hypothesis using an association found between two variables.
- Can collect a lot of data quicker than a lab. experminet, it is more efficient.
- Does not show a 'cause and effect'.
- Many Hypothesis cannot be easily examined, rendering research 'useless'?
- face-to-face conversations based on a question-and-answer format.
- Structured , meaning they are scientific, quantitive and closed questions.
- or Unstructured, meaning they are non-scientific, qualitative and open questions, more like a discussion.
- Qualititive, rather than Quantitative in their content.
- Good Reliability
- Easily analysable, if structured.
- Little relevence to research,
- Data difficult to analyise, if Unstructured.
- Interviewer Bias: Personalities could change course of interview.
- Watching and reccording behaviour without interfering.
- Observations occur in a natural setting, they don't know they are being observed.
- Allows investigation where other methods aren't possible, perhaps as results cannot be numerically recorded.
- Researcher Bias: Researched cannot remain unbiased. May have favourites, and one person's view will differ to another's.
- Small-Scale replication is difficult.
- Lack of controlled variables