research methods Flashcards
(50 cards)
directional hypothesis
The researcher makes clear the difference that is anticipated and use clear words such as higher/lower
nondirectional hypothesis
Simple states that there will be a difference. But the nature of the difference is not specified, such as
“People who do drink the energy drink will differ to those who don’t”
extranous variables
The key thing to an experiment is the variables, such as
The independent variable is changed and manipulated
The dependent variable is what you’re measuring and finding out
The only thing changing the dependent variable is the independent variable, but other things that may affect these are called extraneous variables these then have to be removed
confounding variable
an effect that can add another independent variable and therefore change the dependent variable due to coincidence and chance
demand characteristics
when particapnts use cues and hints about the experiment to try and figure out whats going on which may lead to a change in behavior. and may imply the please u or screw u technique
randomisation
a way of removing extraneous and confounding variables by randomising everything. to reduce the investigative control on stuff
standardisation
using exactly the same formalized procedure and instructions for all participants in a research study
expriemnt design methods
independent groups
repated measures
matched pairs
Independent groups design methods
This is where two separate groups of participants experience two different conditions of the experiment. Then they can use this and compare the effects of the two independent variables
Independent groups evaluation
- People in the groups are not the same, so results may rely on the individual differences. To deal with this, they may use random allocation
- cheaper as you only have to do it once, and quicker
-order effects don’t affect this as they only do it once
Repeated measures design methods
All participants take part in all conditions of the experiment, so they can compare both conditions as they are using the same people, so the difference in personality won’t affect it
Repeated measures evaluation
- A big issue is that the groups have to do at least two tasks meaning one task could affect the actions in another
-Order effects due to the two tasks could cause boredom and fatigue that might cause a decrease in performance; however, their performance may improve if they are doing a skill-based task, as they have more time with it#
-demand chacterstics they have a higher chance of founding out what is happening so more likely to have demand chactersticcs and employ the please u or screw u effects
-more expensive
matched pairs design methods
pairs of participants are matched on a variable first that may affect the dependent variable, and then they are allocated to group A or B separately, an attempt to control the participant’s variables
matched pairs evaluation
-Order effects and demand characteristics are less likely after only doing one task
- participants’ variable effects are reduced massively, but still not 100% as they are still not the same people
- may be more time-consuming and therefore more money
Laboratory experiment strengths
have a high over-extraneous variables, meaning they can ensure that any effect on the dependent variable is likely to be a result of the independent variable and
meaning it has high internal validity
- good replication due to the high level of control and this is important to check the study wasn’t a one off
Laboratory experiment limitations
may lack generalizability as the lab environment may be more artificial and lack natural elements and inan unfamiliar place, participants may act differently
-low external validity
Furthermore, participants are usually aware of being tested in a lab, increasing the chances of demanding characteristics
Tasks in the lab may be unnatural and lack mundane realism
field experiments strength
have higher mundane realism because the environment is natural, so they may produce behaviour that is more natural and authentic. In addition, they may not even know they are being studied, giving it high external validity
Field experiments limitation
loss of control over extraneous variables, which means the causes between the IV and DV in field studies are a lot harder to establish
also important ethical issues if people don’t know they are being studied and haven’t consented
natural experiment strength
provide opportunities for research that may not have otherwise been undertaken for practical or ethical reasons
have high external validity as they involve real-life studies
natural experiment limitation
Opportunities for these experiments are rare, so there are fewer chances to research
hard to see if the IV affected the DV as setup and other factors could be hard to account for
population
a group of people who are the focus of the researcher interests, from which a smaller sample is drawn from them
random sampling
where all members of the population have an equal chance of being picked, firstly you need a list of the members in the population, then assigning them a number, finally, you can pick numbers out of a hat or use a random number generator and get your selected people
systematic sampling
is when every nTH person is selected, for example, every third house on a street, This is when a list from the population is organized in some form, such as alphabetical order, then every 3rd, 4th, 6th, whatever is picked. The number selected can be made random to reduce bias
stratified sampling
putting people into sub groups from the population, such as age, height, and gender .once picking the subgroup you would use random sampling to pick your selected group to avoid bias and get a more narrowed down selection of people
