RM: Scientific Processes (L5-11) Flashcards
What is an aim?
- a precise statement about the purpose of the study and what it intends to find out
- should include what is being studied and what
the study is trying to achieve
What is a hypothesis?
- a specific, testable statement about the expected
outcome of a study - the hypothesis should also be operationalised
What is the first part of a hypothesis that needs addressing?
- whether or not the study predicts causation or correlation
What is correlation?
- when the researcher predicts a relationship between two variables (co-variables) being investigated
What is causation?
- when the researcher predicts a difference in the DV because of the manipulation of an IV
What are significant differences?
- differences in the DV resulting from manipulation of the IV are known as significant differences
- this is if it has been statistically shown using inferential statistics that the differences are highly unlikely to be due to chance
What are the 2 hypotheses in a study?
- null
- alternative
What is a null hypothesis?
- states that the IV will have no effect on the DV
What is an alternative hypothesis?
- predicts that the IV will have an effect on
the DV
What 2 categories are there for an alternative hypothesis?
- non directional, two tailed
- directional, one tailed
What is a non directional + directional hypothesis?
- nd, DOES NOT state the direction of the
predicted differences between conditions - d, DOES state the direction of the predicted
difference between conditions
What determines whether to use a nd or d hypothesis?
- based on whether there is previous research in
the field - if there is, we use a directional hypothesis
- if not, we use a non-directional hypothesis
- this is because previous research will enable
us to predict which direction the results are likely to go in
What to remember when writing a hypothesis?
- test of causation or correlation?
- null, non-directional or directional hypothesis?
- in the correct tense?
- have all the variables been included?
- have all the variables been operationalised?
What are pilot studies?
- small-scale investigations conducted before research
- useful because they can help to identify whether there needs to be any modifications in the design of the planned study
- they also help to determine whether it would be feasible and worthwhile to conduct a full scale study
What is a target population?
- used to describe the group who researchers are studying and want to generalise their results to
What are sampling techniques used for?
- to obtain a sample of the target population
- essential to avoid studying entire target populations, which would take loo tong and be too expensive
A sample should be…
- representative of the population from which it is drawn
- so that the findings of the study can be generalised to the target population
What is random sampling?
- when every member of the target population has the same chance of being selected
- easiest way to do this is to place all names from the target population in a hat and draw out the sample required
+ve and -ve of random sampling:
+ likely to be representative and therefore results can be generalised to the target population
- sometimes difficult to get full details of a target population from which to select a sample
- not all members of the target population who are selected to take part will be available or willing to take part, making the sample unrepresentative
What is systematic sampling?
- sampling technique where participants are selected by taking every Nth person from a list
+ve and -ve of systematic sampling:
+ far simpler than random sampling
- process of selection can interact with a hidden periodic trait within the target population, if the sampling technique coincides with the periodicity of the trait, the sampling technique will no longer be representative
What is stratified sampling?
- involves classifying the target population into
categories - then randomly choosing a sample that consists of participants from each category in the same proportions as they appear in the target population
+ve and -ve of stratified sampling:
+ all groups within a target pop are included, so the sample should be representative of the target population
+ takes into consideration proportion
- can be very time consuming as the categories have to be identified and calculated
- if you do not have details of all the people in your target population you would struggle to conduct a stratified sample
What is opportunity sampling?
- involves selecting participants who are readily
available and willing to take part
+ve and -ve of opportunity sampling:
+ easiest and most practical method of ensuring large samples
- high chance that the sample will not be representative of the target population
- sometimes people feel obliged to take part in
research even when they do not really want to, this is unethical
What is volunteer sampling?
- involves people self-selecting to participate in a
study - researcher will usually advertise for people to take part in their research
+ve and -ve of volunteering sampling:
+ can be a useful way of finding specific people to take part in particular areas of research
- volunteer bias, certain type of individuals (people who are more confident/helpful/curious) tend to volunteer for research, means that there is a very high chance that the sample obtained will be unrepresentative
What is the experimental design of a study?
- how the participants are assigned
to different conditions
3 main types of experimental designs?
- independent groups
- repeated measures
- matched pairs
What is the independent groups design?
- different ps are used in each of the conditions
- so each group of participants is independent from one another
- ps are usually randomly allocated to each condition to balance out any participant variables
+ve independent groups design:
- order effects will not occur as there are different ps in each condition
- order effects are when the sequence in which
ps take part in conditions influences their performance or behaviour - e.g. in a memory test participants may get better with practice
- ps may also get tired or bored or fatigued when being asked to take part in more than one condition
= chance of demand characteristics is reduced as ps do only one condition each and so have less chance to guess the purpose of the study - same task/materials can be used in both conditions as ps are always naïve to the task
-ve independent groups design:
- more ps are needed for this experimental design
= always a chance that the different results between the two conditions are due to participant variables rather than manipulation of the independent variable (IV)
What is the repeated measures design?
- each participant is tested in all
conditions of the experiment
+ve repeated measures design:
- as the same people are measured in all conditions there are no participant variables between the conditions
= half as many ps are needed compared to an independent groups design
-ve repeated measures design:
- order effects may affect the results
- one way to avoid this is counterbalancing, this is when half the participants do condition A first and condition B second and the other half of the participants do condition B first and condition A second
- counterbalancing does not eliminate order
effects, which will be present because there are two separate tasks to be completed by each person - but counterbalancing controls the
impact of order effects (practice, fatigue or boredom) and allows order effects to be distributed evenly across both conditions
= demand characteristics are more likely to occur as ps are involved in the entire study - design takes more time, especially if a time gap between different conditions is required
What is the matched pairs design?
- different participants are used in all of
the conditions, just as with the independent groups design - however, ps in the two groups are matched on characteristics important for that study
- such as age, gender, level of education etc.
- identical twins are often used in matched pairs designs
+ve matched pairs design:
- less risk of order effects
= less risk of demand characteristics. - participant variables are unlikely as the groups have been closely matched
-ve matched pairs design:
- twice as many participants are required compared with a repeated measures design
= matching process is incredibly difficult; even two closely matched individuals have different levels of motivation and fatigue at any given
time - matching process in incredibly time consuming
In order to ensure that a study has validity…
- extraneous variables must be controlled for to prevent them from becoming confounding variables
What are participant variables?
- characteristics of the participants which may affect the dependent variable (DV)
- e.g. intelligence, age, gender, personality etc
How can participant variables be overcome?
- appropriate experimental design can help to try and overcome these types of EV
- matched pairs and repeated measures can help to avoid participant variables
- but repeated measures can lead to order effects, so counterbalancing should
be used to avoid this - random allocation of participants to conditions (e.g. by drawing names out of a hat) when using independent groups should also ensure that groups are not biased
- but random allocation is not possible for a quasi-experiment
What is the process of random allocation (participant variables) ?
- involves all the participants being
identified either by name or number - names/numbers are put in a container (e.g. a hat) or into a computer
- assign alternate names/numbers
drawn to Condition 1 then Condition 2 and so on until there are the required number in each condition - or set the parameters on the computer for two groups to be randomly generated
What are environmental variables?
- factors in the env where the experiment is conducted that could affect the DV
- e.g. temperature, time of day, lighting, noise etc
- the answer to this is standardisation (i.e. making sure that all the
conditions, materials, and instructions are the same for all participants)
